Skip to content

Same agent Smaller bill. The tools and infrastructure to get more work from your AI. Fewer tokens. Less compute. Lower cost.

[Feat]: Update OpenAI Instrumentation to align with OTel Semconv (#616)

Patcher72f3278d099c
diff --git a/docs/latest/connections/grafanacloud.mdx b/docs/latest/connections/grafanacloud.mdx
index cd2fa18e..b3673547 100644
--- a/docs/latest/connections/grafanacloud.mdx
+++ b/docs/latest/connections/grafanacloud.mdx
@@ -1140,7 +1140,7 @@ To directly send OpenTelemetry metrics and traces generated by OpenLIT SDK from
},
"disableTextWrap": false,
"editorMode": "code",
- "expr": "sum by(gen_ai_type) (gen_ai_requests_total{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
+ "expr": "sum by(GEN_AI_OPERATION) (gen_ai_requests_total{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
"fullMetaSearch": false,
"includeNullMetadata": true,
"instant": false,
diff --git a/docs/latest/connections/oodle.mdx b/docs/latest/connections/oodle.mdx
index 1879f938..81914162 100644
--- a/docs/latest/connections/oodle.mdx
+++ b/docs/latest/connections/oodle.mdx
@@ -1124,7 +1124,7 @@ To directly send OpenTelemetry metrics generated by OpenLIT SDK from your AI App
},
"disableTextWrap": false,
"editorMode": "builder",
- "expr": "sum by(gen_ai_type) (openlit_otel_metrics_gen_ai_total_requests{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
+ "expr": "sum by(GEN_AI_OPERATION) (openlit_otel_metrics_gen_ai_total_requests{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
"fullMetaSearch": false,
"includeNullMetadata": true,
"instant": false,
diff --git a/docs/latest/connections/prometheus-jaeger.mdx b/docs/latest/connections/prometheus-jaeger.mdx
index ce3b0575..dc0eed74 100644
--- a/docs/latest/connections/prometheus-jaeger.mdx
+++ b/docs/latest/connections/prometheus-jaeger.mdx
@@ -1056,7 +1056,7 @@ To directly send OpenTelemetry metrics and traces generated by OpenLIT SDK from
},
"disableTextWrap": false,
"editorMode": "code",
- "expr": "sum by(gen_ai_type) (gen_ai_total_requests{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
+ "expr": "sum by(GEN_AI_OPERATION) (gen_ai_total_requests{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
"fullMetaSearch": false,
"includeNullMetadata": true,
"instant": false,
diff --git a/docs/latest/connections/prometheus-tempo.mdx b/docs/latest/connections/prometheus-tempo.mdx
index 6fd5027a..de72e490 100644
--- a/docs/latest/connections/prometheus-tempo.mdx
+++ b/docs/latest/connections/prometheus-tempo.mdx
@@ -1054,7 +1054,7 @@ To directly send OpenTelemetry metrics and traces generated by OpenLIT SDK from
},
"disableTextWrap": false,
"editorMode": "code",
- "expr": "sum by(gen_ai_type) (gen_ai_total_requests{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
+ "expr": "sum by(GEN_AI_OPERATION) (gen_ai_total_requests{telemetry_sdk_name=\"openlit\", gen_ai_application_name=~\"$application\", gen_ai_environment=~\"$environment\"})",
"fullMetaSearch": false,
"includeNullMetadata": true,
"instant": false,
diff --git a/sdk/python/.pylintrc b/sdk/python/.pylintrc
index 2895b9de..9542de74 100644
--- a/sdk/python/.pylintrc
+++ b/sdk/python/.pylintrc
@@ -1,10 +1,27 @@
[DESIGN]
# Maximum number of locals for function / method body
-max-locals=25
+max-locals=30
# Maximum number of arguments for function / method
max-args=25
# Maximum number of positional arguments (see R0917).
max-positional-arguments=12
+
+# Maximum allowed line length
+max-line-length=120
+
+# Maximum number of instance attributes
+max-attributes=20
+
+[MESSAGES CONTROL]
+disable=
+ global-statement,
+ duplicate-code,
+ broad-exception-caught,
+ too-many-statements,
+ unused-argument,
+ too-many-branches,
+ too-many-locals,
+ too-many-lines
\ No newline at end of file
diff --git a/sdk/python/pyproject.toml b/sdk/python/pyproject.toml
index 9e20da8b..dde0383f 100644
--- a/sdk/python/pyproject.toml
+++ b/sdk/python/pyproject.toml
@@ -1,6 +1,6 @@
[tool.poetry]
name = "openlit"
-version = "1.33.8"
+version = "1.33.9"
description = "OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications and GPUs, facilitating the integration of observability into your GenAI-driven projects"
authors = ["OpenLIT"]
license = "Apache-2.0"
diff --git a/sdk/python/src/openlit/__helpers.py b/sdk/python/src/openlit/__helpers.py
index e71c01f6..626b56aa 100644
--- a/sdk/python/src/openlit/__helpers.py
+++ b/sdk/python/src/openlit/__helpers.py
@@ -6,9 +6,12 @@ import os
import json
import logging
from urllib.parse import urlparse
+from typing import Any, Dict, List
import requests
import tiktoken
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
from opentelemetry.trace import Status, StatusCode
+from openlit.semcov import SemanticConvetion
# Set up logging
logger = logging.getLogger(__name__)
@@ -197,3 +200,52 @@ def handle_exception(span,e):
# Record the exception details within the span
span.record_exception(e)
span.set_status(Status(StatusCode.ERROR))
+
+def calculate_ttft(timestamps: List[float], start_time: float) -> float:
+ """
+ Calculate the time to the first tokens.
+
+ :param timestamps: List of timestamps for received tokens
+ :param start_time: The start time of the streaming process
+ :return: Time to the first tokens
+ """
+ if timestamps:
+ return timestamps[0] - start_time
+ return 0.0
+
+def calculate_tbt(timestamps: List[float]) -> float:
+ """
+ Calculate the average time between tokens.
+
+ :param timestamps: List of timestamps for received tokens
+ :return: Average time between tokens
+ """
+ if len(timestamps) > 1:
+ time_diffs = [timestamps[i] - timestamps[i - 1] for i in range(1, len(timestamps))]
+ return sum(time_diffs) / len(time_diffs)
+ return 0.0
+
+def create_metrics_attributes(
+ service_name: str,
+ deployment_environment: str,
+ operation: str,
+ system: str,
+ request_model: str,
+ server_address: str,
+ server_port: int,
+ response_model: str,
+) -> Dict[Any, Any]:
+ """
+ Returns OTel metrics attributes
+ """
+ return {
+ TELEMETRY_SDK_NAME: "openlit",
+ SERVICE_NAME: service_name,
+ DEPLOYMENT_ENVIRONMENT: deployment_environment,
+ SemanticConvetion.GEN_AI_OPERATION: operation,
+ SemanticConvetion.GEN_AI_SYSTEM: system,
+ SemanticConvetion.GEN_AI_REQUEST_MODEL: request_model,
+ SemanticConvetion.SERVER_ADDRESS: server_address,
+ SemanticConvetion.SERVER_PORT: server_port,
+ SemanticConvetion.GEN_AI_RESPONSE_MODEL: response_model
+ }
diff --git a/sdk/python/src/openlit/instrumentation/ag2/ag2.py b/sdk/python/src/openlit/instrumentation/ag2/ag2.py
index 28d638c2..69e42385 100644
--- a/sdk/python/src/openlit/instrumentation/ag2/ag2.py
+++ b/sdk/python/src/openlit/instrumentation/ag2/ag2.py
@@ -74,8 +74,8 @@ def wrap_ag2(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_AGENT_ROLE,
instance.name)
if llm_model:
diff --git a/sdk/python/src/openlit/instrumentation/ai21/ai21.py b/sdk/python/src/openlit/instrumentation/ai21/ai21.py
index 560f0553..dacd4b14 100644
--- a/sdk/python/src/openlit/instrumentation/ai21/ai21.py
+++ b/sdk/python/src/openlit/instrumentation/ai21/ai21.py
@@ -126,8 +126,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AI21)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -146,9 +146,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
self._kwargs.get("temperature", 1.0))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
self._prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
self._completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
self._prompt_tokens + self._completion_tokens)
@@ -180,8 +180,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AI21,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
self._kwargs.get("model", "jamba-1.5-mini")
}
@@ -261,8 +261,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AI21)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -296,9 +296,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
pricing_info, response_dict.get('usage', {}).get('prompt_tokens', None),
response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage', {}).get('completion_tokens', None))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage', {}).get('total_tokens', None))
@@ -345,9 +345,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage').get('completion_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -366,8 +366,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AI21,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "jamba-1.5-mini")
}
@@ -470,8 +470,8 @@ def chat_rag(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AI21)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -553,9 +553,9 @@ def chat_rag(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -579,8 +579,8 @@ def chat_rag(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AI21,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "jamba-1.5-mini")
}
diff --git a/sdk/python/src/openlit/instrumentation/ai21/async_ai21.py b/sdk/python/src/openlit/instrumentation/ai21/async_ai21.py
index 3b0a0dff..07675a40 100644
--- a/sdk/python/src/openlit/instrumentation/ai21/async_ai21.py
+++ b/sdk/python/src/openlit/instrumentation/ai21/async_ai21.py
@@ -126,8 +126,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AI21)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -146,9 +146,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
self._kwargs.get("temperature", 1.0))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
self._prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
self._completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
self._prompt_tokens + self._completion_tokens)
@@ -180,8 +180,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AI21,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
self._kwargs.get("model", "jamba-1.5-mini")
}
@@ -260,8 +260,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AI21)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -295,9 +295,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
pricing_info, response_dict.get('usage', {}).get('prompt_tokens', None),
response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage', {}).get('completion_tokens', None))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage', {}).get('total_tokens', None))
@@ -344,9 +344,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage').get('completion_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -365,8 +365,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AI21,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "jamba-1.5-mini")
}
@@ -469,8 +469,8 @@ def async_chat_rag(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AI21)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -552,9 +552,9 @@ def async_chat_rag(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -578,8 +578,8 @@ def async_chat_rag(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AI21,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "jamba-1.5-mini")
}
diff --git a/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py b/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py
index 4cda8573..037e77b1 100644
--- a/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py
+++ b/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py
@@ -107,8 +107,8 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -131,9 +131,9 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -165,8 +165,8 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "claude-3-sonnet-20240229")
}
@@ -219,8 +219,8 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -243,9 +243,9 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[response.stop_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.input_tokens +
@@ -280,8 +280,8 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "claude-3-sonnet-20240229")
}
diff --git a/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py b/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py
index 4f822141..7b6bd01c 100644
--- a/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py
+++ b/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py
@@ -107,8 +107,8 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -131,9 +131,9 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -165,8 +165,8 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "claude-3-sonnet-20240229")
}
@@ -219,8 +219,8 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -243,9 +243,9 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[response.stop_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.input_tokens +
@@ -279,8 +279,8 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_ANTHROPIC,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "claude-3-sonnet-20240229")
}
diff --git a/sdk/python/src/openlit/instrumentation/assemblyai/assemblyai.py b/sdk/python/src/openlit/instrumentation/assemblyai/assemblyai.py
index b248b9b1..2bdfdf96 100644
--- a/sdk/python/src/openlit/instrumentation/assemblyai/assemblyai.py
+++ b/sdk/python/src/openlit/instrumentation/assemblyai/assemblyai.py
@@ -62,8 +62,8 @@ def transcribe(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_ASSEMBLYAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AUDIO)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -105,8 +105,8 @@ def transcribe(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_ASSEMBLYAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_AUDIO,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
llm_model,
}
diff --git a/sdk/python/src/openlit/instrumentation/astra/astra.py b/sdk/python/src/openlit/instrumentation/astra/astra.py
index 655e38d0..a125b03c 100644
--- a/sdk/python/src/openlit/instrumentation/astra/astra.py
+++ b/sdk/python/src/openlit/instrumentation/astra/astra.py
@@ -73,8 +73,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_VECTORDB)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
span.set_attribute(SemanticConvetion.DB_SYSTEM,
SemanticConvetion.DB_SYSTEM_ASTRA)
@@ -205,8 +205,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.DB_SYSTEM_ASTRA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_VECTORDB,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,
SemanticConvetion.DB_OPERATION:
db_operation
}
diff --git a/sdk/python/src/openlit/instrumentation/astra/async_astra.py b/sdk/python/src/openlit/instrumentation/astra/async_astra.py
index 2f93bba2..71da0c87 100644
--- a/sdk/python/src/openlit/instrumentation/astra/async_astra.py
+++ b/sdk/python/src/openlit/instrumentation/astra/async_astra.py
@@ -73,8 +73,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_VECTORDB)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
span.set_attribute(SemanticConvetion.DB_SYSTEM,
SemanticConvetion.DB_SYSTEM_ASTRA)
@@ -205,8 +205,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.DB_SYSTEM_ASTRA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_VECTORDB,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,
SemanticConvetion.DB_OPERATION:
db_operation
}
diff --git a/sdk/python/src/openlit/instrumentation/azure_ai_inference/async_azure_ai_inference.py b/sdk/python/src/openlit/instrumentation/azure_ai_inference/async_azure_ai_inference.py
index 7a4f2124..e22b638d 100644
--- a/sdk/python/src/openlit/instrumentation/azure_ai_inference/async_azure_ai_inference.py
+++ b/sdk/python/src/openlit/instrumentation/azure_ai_inference/async_azure_ai_inference.py
@@ -107,8 +107,8 @@ def async_complete(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -135,9 +135,9 @@ def async_complete(gen_ai_endpoint, version, environment, application_name,
kwargs.get("seed", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response_id)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -171,8 +171,8 @@ def async_complete(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -221,8 +221,8 @@ def async_complete(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -273,9 +273,9 @@ def async_complete(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(model,
pricing_info, input_tokens, output_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -294,8 +294,8 @@ def async_complete(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -365,8 +365,8 @@ def async_embedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -375,13 +375,13 @@ def async_embedding(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get("model", "text-embedding-ada-002"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("encoding_format", "float"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
kwargs.get("dimensions", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -407,8 +407,8 @@ def async_embedding(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "text-embedding-ada-002")
}
diff --git a/sdk/python/src/openlit/instrumentation/azure_ai_inference/azure_ai_inference.py b/sdk/python/src/openlit/instrumentation/azure_ai_inference/azure_ai_inference.py
index 17b2ebda..2612afc8 100644
--- a/sdk/python/src/openlit/instrumentation/azure_ai_inference/azure_ai_inference.py
+++ b/sdk/python/src/openlit/instrumentation/azure_ai_inference/azure_ai_inference.py
@@ -107,8 +107,8 @@ def complete(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -135,9 +135,9 @@ def complete(gen_ai_endpoint, version, environment, application_name,
kwargs.get("seed", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response_id)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -171,8 +171,8 @@ def complete(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -221,8 +221,8 @@ def complete(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -273,9 +273,9 @@ def complete(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(model,
pricing_info, input_tokens, output_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -294,8 +294,8 @@ def complete(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -365,8 +365,8 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -375,13 +375,13 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get("model", "text-embedding-ada-002"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("encoding_format", "float"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
kwargs.get("dimensions", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -407,8 +407,8 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_AZURE_AI_INFERENCE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "text-embedding-ada-002")
}
diff --git a/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py b/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py
index ae5ec582..281dd991 100644
--- a/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py
+++ b/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py
@@ -115,7 +115,7 @@ def converse(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_BEDROCK)
+ SemanticConvetion.GEN_AI_SYSTEM_AWS_BEDROCK)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -124,9 +124,9 @@ def converse(gen_ai_endpoint, version, environment, application_name, tracer,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
model)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
input_tokens + output_tokens)
@@ -162,11 +162,11 @@ def converse(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_APPLICATION_NAME:
application_name,
SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_BEDROCK,
+ SemanticConvetion.GEN_AI_SYSTEM_AWS_BEDROCK,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/instrumentation/chroma/chroma.py b/sdk/python/src/openlit/instrumentation/chroma/chroma.py
index 34e5974d..a853a218 100644
--- a/sdk/python/src/openlit/instrumentation/chroma/chroma.py
+++ b/sdk/python/src/openlit/instrumentation/chroma/chroma.py
@@ -77,8 +77,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_VECTORDB)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
span.set_attribute(SemanticConvetion.DB_SYSTEM,
SemanticConvetion.DB_SYSTEM_CHROMA)
span.set_attribute(SemanticConvetion.DB_COLLECTION_NAME,
@@ -179,8 +179,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.DB_SYSTEM_CHROMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_VECTORDB,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,
SemanticConvetion.DB_OPERATION:
db_operation
}
diff --git a/sdk/python/src/openlit/instrumentation/cohere/cohere.py b/sdk/python/src/openlit/instrumentation/cohere/cohere.py
index b96b3c07..b78bd9db 100644
--- a/sdk/python/src/openlit/instrumentation/cohere/cohere.py
+++ b/sdk/python/src/openlit/instrumentation/cohere/cohere.py
@@ -64,8 +64,8 @@ def embed(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_COHERE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -74,7 +74,7 @@ def embed(gen_ai_endpoint, version, environment, application_name, tracer,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get("model", "embed-english-v2.0"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("embedding_types", "float"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
kwargs.get("input_type", ""))
@@ -82,7 +82,7 @@ def embed(gen_ai_endpoint, version, environment, application_name, tracer,
kwargs.get("user", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.meta.billed_units.input_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.meta.billed_units.input_tokens)
@@ -108,8 +108,8 @@ def embed(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_SYSTEM_COHERE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "embed-english-v2.0")
}
@@ -184,8 +184,8 @@ def chat(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_COHERE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -210,9 +210,9 @@ def chat(gen_ai_endpoint, version, environment, application_name, tracer,
response.generation_id)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[response.finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.meta.billed_units.input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.meta.billed_units.output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.meta.billed_units.input_tokens +
@@ -246,8 +246,8 @@ def chat(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_SYSTEM_COHERE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "command")
}
@@ -336,8 +336,8 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_COHERE)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -362,9 +362,9 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
response_id)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -396,8 +396,8 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_COHERE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "command")
}
diff --git a/sdk/python/src/openlit/instrumentation/controlflow/controlflow.py b/sdk/python/src/openlit/instrumentation/controlflow/controlflow.py
index 878307ab..8848586b 100644
--- a/sdk/python/src/openlit/instrumentation/controlflow/controlflow.py
+++ b/sdk/python/src/openlit/instrumentation/controlflow/controlflow.py
@@ -68,8 +68,8 @@ def wrap_controlflow(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
if gen_ai_endpoint == "controlflow.create_agent":
span.set_attribute(SemanticConvetion.GEN_AI_AGENT_ROLE,
diff --git a/sdk/python/src/openlit/instrumentation/crawl4ai/async_crawl4ai.py b/sdk/python/src/openlit/instrumentation/crawl4ai/async_crawl4ai.py
index eab21a59..b2aa4fbb 100644
--- a/sdk/python/src/openlit/instrumentation/crawl4ai/async_crawl4ai.py
+++ b/sdk/python/src/openlit/instrumentation/crawl4ai/async_crawl4ai.py
@@ -58,8 +58,8 @@ def async_wrap_crawl(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_CRAWL4AI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/crawl4ai/crawl4ai.py b/sdk/python/src/openlit/instrumentation/crawl4ai/crawl4ai.py
index 82b67e4c..97ecc04d 100644
--- a/sdk/python/src/openlit/instrumentation/crawl4ai/crawl4ai.py
+++ b/sdk/python/src/openlit/instrumentation/crawl4ai/crawl4ai.py
@@ -58,8 +58,8 @@ def wrap_crawl(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_CRAWL4AI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/crewai/crewai.py b/sdk/python/src/openlit/instrumentation/crewai/crewai.py
index 8d4b7b14..b6e88288 100644
--- a/sdk/python/src/openlit/instrumentation/crewai/crewai.py
+++ b/sdk/python/src/openlit/instrumentation/crewai/crewai.py
@@ -71,8 +71,8 @@ def crew_wrap(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_CREWAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/dynamiq/dynamiq.py b/sdk/python/src/openlit/instrumentation/dynamiq/dynamiq.py
index 62ac21f8..3de1a637 100644
--- a/sdk/python/src/openlit/instrumentation/dynamiq/dynamiq.py
+++ b/sdk/python/src/openlit/instrumentation/dynamiq/dynamiq.py
@@ -58,8 +58,8 @@ def dynamiq_wrap(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_DYNAMIQ)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py b/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py
index 36a8b586..529a6817 100644
--- a/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py
+++ b/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py
@@ -60,8 +60,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AUDIO)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -99,8 +99,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_AUDIO,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "eleven_multilingual_v2")
}
diff --git a/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py b/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py
index 536a0b8b..b4df23c0 100644
--- a/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py
+++ b/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py
@@ -60,8 +60,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AUDIO)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -105,8 +105,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_ELEVENLABS,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_AUDIO,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "eleven_multilingual_v2")
}
diff --git a/sdk/python/src/openlit/instrumentation/embedchain/embedchain.py b/sdk/python/src/openlit/instrumentation/embedchain/embedchain.py
index 80d931f3..95765a00 100644
--- a/sdk/python/src/openlit/instrumentation/embedchain/embedchain.py
+++ b/sdk/python/src/openlit/instrumentation/embedchain/embedchain.py
@@ -65,8 +65,8 @@ def evaluate(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_EVAL_CONTEXT_RELEVANCY,
@@ -143,8 +143,8 @@ def get_data_sources(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_DATA_SOURCES,
diff --git a/sdk/python/src/openlit/instrumentation/firecrawl/firecrawl.py b/sdk/python/src/openlit/instrumentation/firecrawl/firecrawl.py
index 9e0f2f66..53f137a6 100644
--- a/sdk/python/src/openlit/instrumentation/firecrawl/firecrawl.py
+++ b/sdk/python/src/openlit/instrumentation/firecrawl/firecrawl.py
@@ -58,8 +58,8 @@ def wrap_crawl(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_FIRECRAWL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/google_ai_studio/async_google_ai_studio.py b/sdk/python/src/openlit/instrumentation/google_ai_studio/async_google_ai_studio.py
index 8f456a8c..943453ad 100644
--- a/sdk/python/src/openlit/instrumentation/google_ai_studio/async_google_ai_studio.py
+++ b/sdk/python/src/openlit/instrumentation/google_ai_studio/async_google_ai_studio.py
@@ -94,8 +94,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -106,9 +106,9 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -140,8 +140,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -183,8 +183,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -217,9 +217,9 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(model,
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -238,8 +238,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/instrumentation/google_ai_studio/google_ai_studio.py b/sdk/python/src/openlit/instrumentation/google_ai_studio/google_ai_studio.py
index e531d24b..3c2eb8ec 100644
--- a/sdk/python/src/openlit/instrumentation/google_ai_studio/google_ai_studio.py
+++ b/sdk/python/src/openlit/instrumentation/google_ai_studio/google_ai_studio.py
@@ -94,8 +94,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -106,9 +106,9 @@ def generate(gen_ai_endpoint, version, environment, application_name,
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -140,8 +140,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -184,8 +184,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -218,9 +218,9 @@ def generate(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(model,
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -239,8 +239,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GOOGLE_AI_STUDIO,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py b/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py
index 9c6b5494..5101bca5 100644
--- a/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py
+++ b/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py
@@ -84,8 +84,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GPT4ALL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -120,9 +120,9 @@ def generate(gen_ai_endpoint, version, environment, application_name,
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -141,8 +141,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GPT4ALL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -180,8 +180,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GPT4ALL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -216,9 +216,9 @@ def generate(gen_ai_endpoint, version, environment, application_name,
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -237,8 +237,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GPT4ALL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -312,8 +312,8 @@ def embed(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GPT4ALL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -322,7 +322,7 @@ def embed(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
model)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens)
@@ -348,8 +348,8 @@ def embed(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GPT4ALL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/instrumentation/groq/async_groq.py b/sdk/python/src/openlit/instrumentation/groq/async_groq.py
index 48434c88..1033d418 100644
--- a/sdk/python/src/openlit/instrumentation/groq/async_groq.py
+++ b/sdk/python/src/openlit/instrumentation/groq/async_groq.py
@@ -104,8 +104,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GROQ)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -132,9 +132,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
kwargs.get("seed", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -166,8 +166,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GROQ,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "gpt-3.5-turbo")
}
@@ -216,8 +216,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GROQ)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -259,9 +259,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
pricing_info, response.usage.prompt_tokens,
response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -305,9 +305,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
"Function called with tools")
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -326,8 +326,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GROQ,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "gpt-3.5-turbo")
}
diff --git a/sdk/python/src/openlit/instrumentation/groq/groq.py b/sdk/python/src/openlit/instrumentation/groq/groq.py
index d057d937..df89f15d 100644
--- a/sdk/python/src/openlit/instrumentation/groq/groq.py
+++ b/sdk/python/src/openlit/instrumentation/groq/groq.py
@@ -104,8 +104,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GROQ)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -132,9 +132,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
kwargs.get("seed", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -166,8 +166,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GROQ,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "gpt-3.5-turbo")
}
@@ -216,8 +216,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_GROQ)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -259,9 +259,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
pricing_info, response.usage.prompt_tokens,
response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -305,9 +305,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
"Function called with tools")
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -326,8 +326,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_GROQ,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "gpt-3.5-turbo")
}
diff --git a/sdk/python/src/openlit/instrumentation/haystack/haystack.py b/sdk/python/src/openlit/instrumentation/haystack/haystack.py
index e0479bb1..56154fc9 100644
--- a/sdk/python/src/openlit/instrumentation/haystack/haystack.py
+++ b/sdk/python/src/openlit/instrumentation/haystack/haystack.py
@@ -65,8 +65,8 @@ def join_data(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/julep/async_julep.py b/sdk/python/src/openlit/instrumentation/julep/async_julep.py
index 67936d62..54de6ab4 100644
--- a/sdk/python/src/openlit/instrumentation/julep/async_julep.py
+++ b/sdk/python/src/openlit/instrumentation/julep/async_julep.py
@@ -68,8 +68,8 @@ def async_wrap_julep(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
if gen_ai_endpoint == "julep.agents_create":
span.set_attribute(SemanticConvetion.GEN_AI_AGENT_ID,
diff --git a/sdk/python/src/openlit/instrumentation/julep/julep.py b/sdk/python/src/openlit/instrumentation/julep/julep.py
index 9bb02aed..cb02b2e4 100644
--- a/sdk/python/src/openlit/instrumentation/julep/julep.py
+++ b/sdk/python/src/openlit/instrumentation/julep/julep.py
@@ -68,8 +68,8 @@ def wrap_julep(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
if gen_ai_endpoint == "julep.agents_create":
span.set_attribute(SemanticConvetion.GEN_AI_AGENT_ID,
diff --git a/sdk/python/src/openlit/instrumentation/langchain/langchain.py b/sdk/python/src/openlit/instrumentation/langchain/langchain.py
index 399234ca..65b5bd9a 100644
--- a/sdk/python/src/openlit/instrumentation/langchain/langchain.py
+++ b/sdk/python/src/openlit/instrumentation/langchain/langchain.py
@@ -85,8 +85,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_RETRIEVAL_SOURCE,
@@ -159,8 +159,8 @@ def hub(gen_ai_endpoint, version, environment, application_name, tracer,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_HUB_OWNER,
@@ -247,8 +247,8 @@ def allm(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
@@ -261,9 +261,9 @@ def allm(gen_ai_endpoint, version, environment, application_name,
str(get_attribute_from_instance_or_kwargs(instance, 'top_p')))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
input_tokens + output_tokens)
@@ -295,8 +295,8 @@ def allm(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LANGCHAIN,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
str(get_attribute_from_instance_or_kwargs(instance, 'model'))
}
@@ -387,8 +387,8 @@ def llm(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
@@ -401,9 +401,9 @@ def llm(gen_ai_endpoint, version, environment, application_name,
str(get_attribute_from_instance_or_kwargs(instance, 'top_p')))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
input_tokens + output_tokens)
@@ -435,8 +435,8 @@ def llm(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LANGCHAIN,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
str(get_attribute_from_instance_or_kwargs(instance, 'model'))
}
@@ -540,8 +540,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
@@ -554,9 +554,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
str(getattr(instance, 'top_p', 1)))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
input_tokens + output_tokens)
@@ -589,8 +589,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LANGCHAIN,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -695,8 +695,8 @@ def achat(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
@@ -709,9 +709,9 @@ def achat(gen_ai_endpoint, version, environment, application_name,
str(getattr(instance, 'top_p',1)))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
input_tokens + output_tokens)
@@ -743,8 +743,8 @@ def achat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LANGCHAIN,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/instrumentation/letta/letta.py b/sdk/python/src/openlit/instrumentation/letta/letta.py
index f99ca0b9..e0058afa 100644
--- a/sdk/python/src/openlit/instrumentation/letta/letta.py
+++ b/sdk/python/src/openlit/instrumentation/letta/letta.py
@@ -58,8 +58,8 @@ def create_agent(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_LETTA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
@@ -139,17 +139,17 @@ def send_message(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_LETTA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_AGENT_STEP_COUNT,
response.usage.step_count)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
diff --git a/sdk/python/src/openlit/instrumentation/litellm/async_litellm.py b/sdk/python/src/openlit/instrumentation/litellm/async_litellm.py
index 57a978cc..646e9318 100644
--- a/sdk/python/src/openlit/instrumentation/litellm/async_litellm.py
+++ b/sdk/python/src/openlit/instrumentation/litellm/async_litellm.py
@@ -126,8 +126,8 @@ def acompletion(gen_ai_endpoint, version, environment, application_name,
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -154,9 +154,9 @@ def acompletion(gen_ai_endpoint, version, environment, application_name,
self._kwargs.get("seed", ""))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -188,8 +188,8 @@ def acompletion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
self._kwargs.get("model", "gpt-3.5-turbo")
}
@@ -269,8 +269,8 @@ def acompletion(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -312,9 +312,9 @@ def acompletion(gen_ai_endpoint, version, environment, application_name,
pricing_info, response_dict.get('usage', {}).get('prompt_tokens', None),
response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage', {}).get('completion_tokens', None))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage', {}).get('total_tokens', None))
@@ -361,9 +361,9 @@ def acompletion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage').get('completion_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -382,8 +382,8 @@ def acompletion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "gpt-3.5-turbo")
}
@@ -453,8 +453,8 @@ def aembedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -463,13 +463,13 @@ def aembedding(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get("model", "text-embedding-ada-002"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("encoding_format", "float"))
# span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
# kwargs.get("dimensions", "null"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -495,8 +495,8 @@ def aembedding(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "text-embedding-ada-002")
}
diff --git a/sdk/python/src/openlit/instrumentation/litellm/litellm.py b/sdk/python/src/openlit/instrumentation/litellm/litellm.py
index af038a83..1f233a66 100644
--- a/sdk/python/src/openlit/instrumentation/litellm/litellm.py
+++ b/sdk/python/src/openlit/instrumentation/litellm/litellm.py
@@ -126,8 +126,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -154,9 +154,9 @@ def completion(gen_ai_endpoint, version, environment, application_name,
self._kwargs.get("seed", ""))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -188,8 +188,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
self._kwargs.get("model", "gpt-3.5-turbo")
}
@@ -269,8 +269,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -312,9 +312,9 @@ def completion(gen_ai_endpoint, version, environment, application_name,
pricing_info, response_dict.get('usage', {}).get('prompt_tokens', None),
response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage', {}).get('completion_tokens', None))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage', {}).get('total_tokens', None))
@@ -361,9 +361,9 @@ def completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage').get('completion_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -382,8 +382,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_LITELLM,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "gpt-3.5-turbo")
}
@@ -453,8 +453,8 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -463,13 +463,13 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get("model", "text-embedding-ada-002"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("encoding_format", "float"))
# span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
# kwargs.get("dimensions", "null"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -495,8 +495,8 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "text-embedding-ada-002")
}
diff --git a/sdk/python/src/openlit/instrumentation/llamaindex/llamaindex.py b/sdk/python/src/openlit/instrumentation/llamaindex/llamaindex.py
index 40e0e000..0468172f 100644
--- a/sdk/python/src/openlit/instrumentation/llamaindex/llamaindex.py
+++ b/sdk/python/src/openlit/instrumentation/llamaindex/llamaindex.py
@@ -65,8 +65,8 @@ def load_data(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_RETRIEVAL_SOURCE,
diff --git a/sdk/python/src/openlit/instrumentation/mem0/mem0.py b/sdk/python/src/openlit/instrumentation/mem0/mem0.py
index 90298caf..82b101cf 100644
--- a/sdk/python/src/openlit/instrumentation/mem0/mem0.py
+++ b/sdk/python/src/openlit/instrumentation/mem0/mem0.py
@@ -65,8 +65,8 @@ def mem0_wrap(gen_ai_endpoint, version, environment, application_name,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FRAMEWORK)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_FRAMEWORK)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
diff --git a/sdk/python/src/openlit/instrumentation/milvus/milvus.py b/sdk/python/src/openlit/instrumentation/milvus/milvus.py
index dea56b5b..c26b6591 100644
--- a/sdk/python/src/openlit/instrumentation/milvus/milvus.py
+++ b/sdk/python/src/openlit/instrumentation/milvus/milvus.py
@@ -77,8 +77,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_VECTORDB)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
span.set_attribute(SemanticConvetion.DB_SYSTEM,
SemanticConvetion.DB_SYSTEM_MILVUS)
@@ -159,8 +159,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.DB_SYSTEM_MILVUS,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_VECTORDB,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,
SemanticConvetion.DB_OPERATION:
db_operation
}
diff --git a/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py b/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py
index 2673123a..a8254d94 100644
--- a/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py
+++ b/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py
@@ -80,8 +80,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -104,9 +104,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
False)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[response.choices[0].finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -139,8 +139,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "mistral-small-latest")
}
@@ -246,8 +246,8 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -270,9 +270,9 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
True)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -304,8 +304,8 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "mistral-small-latest")
}
@@ -374,8 +374,8 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -384,11 +384,11 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get('model', "mistral-embed"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("encoding_format", "float"))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -414,8 +414,8 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get('model', "mistral-embed")
}
diff --git a/sdk/python/src/openlit/instrumentation/mistral/mistral.py b/sdk/python/src/openlit/instrumentation/mistral/mistral.py
index d00410ee..dfd48023 100644
--- a/sdk/python/src/openlit/instrumentation/mistral/mistral.py
+++ b/sdk/python/src/openlit/instrumentation/mistral/mistral.py
@@ -79,8 +79,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -103,9 +103,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
False)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[response.choices[0].finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -138,8 +138,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "mistral-small-latest")
}
@@ -245,8 +245,8 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -269,9 +269,9 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
True)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
[finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -303,8 +303,8 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "mistral-small-latest")
}
@@ -373,8 +373,8 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -383,11 +383,11 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get('model', "mistral-embed"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("encoding_format", "float"))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage.total_tokens)
@@ -413,8 +413,8 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_MISTRAL,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get('model', "mistral-embed")
}
diff --git a/sdk/python/src/openlit/instrumentation/multion/async_multion.py b/sdk/python/src/openlit/instrumentation/multion/async_multion.py
index 683b76b6..5d7f2f62 100644
--- a/sdk/python/src/openlit/instrumentation/multion/async_multion.py
+++ b/sdk/python/src/openlit/instrumentation/multion/async_multion.py
@@ -58,8 +58,8 @@ def async_multion_wrap(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MULTION)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/multion/multion.py b/sdk/python/src/openlit/instrumentation/multion/multion.py
index fad67349..57c3823b 100644
--- a/sdk/python/src/openlit/instrumentation/multion/multion.py
+++ b/sdk/python/src/openlit/instrumentation/multion/multion.py
@@ -58,8 +58,8 @@ def multion_wrap(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_MULTION)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py b/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py
index dfa8f8cc..df07ad51 100644
--- a/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py
+++ b/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py
@@ -104,8 +104,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -116,9 +116,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
kwargs.get("model", "llama3"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -150,8 +150,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -198,8 +198,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -231,9 +231,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(kwargs.get("model", "llama3"),
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -254,8 +254,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -347,8 +347,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -359,9 +359,9 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
kwargs.get("model", "llama3"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -394,8 +394,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -423,8 +423,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -456,9 +456,9 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(kwargs.get("model", "llama3"),
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -479,8 +479,8 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -550,8 +550,8 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -560,7 +560,7 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get('model', "llama3"))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens)
@@ -586,8 +586,8 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get('model', "llama3")
}
diff --git a/sdk/python/src/openlit/instrumentation/ollama/ollama.py b/sdk/python/src/openlit/instrumentation/ollama/ollama.py
index 49a8ed8c..80b0150b 100644
--- a/sdk/python/src/openlit/instrumentation/ollama/ollama.py
+++ b/sdk/python/src/openlit/instrumentation/ollama/ollama.py
@@ -105,8 +105,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -117,9 +117,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
kwargs.get("model", "llama3"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -151,8 +151,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -199,8 +199,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -232,9 +232,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(kwargs.get("model", "llama3"),
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -255,8 +255,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -348,8 +348,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -360,9 +360,9 @@ def generate(gen_ai_endpoint, version, environment, application_name,
kwargs.get("model", "llama3"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -395,8 +395,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -424,8 +424,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -457,9 +457,9 @@ def generate(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(kwargs.get("model", "llama3"),
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -480,8 +480,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "llama3")
}
@@ -551,8 +551,8 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -561,7 +561,7 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get('model', "llama3"))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens)
@@ -587,8 +587,8 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_OLLAMA,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get('model', "llama3")
}
diff --git a/sdk/python/src/openlit/instrumentation/openai/__init__.py b/sdk/python/src/openlit/instrumentation/openai/__init__.py
index cc4155f2..112675d6 100644
--- a/sdk/python/src/openlit/instrumentation/openai/__init__.py
+++ b/sdk/python/src/openlit/instrumentation/openai/__init__.py
@@ -5,20 +5,12 @@ import importlib.metadata
from opentelemetry.instrumentation.instrumentor import BaseInstrumentor
from wrapt import wrap_function_wrapper
-from openlit.instrumentation.openai.openai import chat_completions, embedding, finetune
+from openlit.instrumentation.openai.openai import chat_completions, embedding
from openlit.instrumentation.openai.openai import image_generate, image_variatons, audio_create
from openlit.instrumentation.openai.async_openai import async_chat_completions, async_embedding
-from openlit.instrumentation.openai.async_openai import async_finetune
from openlit.instrumentation.openai.async_openai import async_image_generate, async_image_variatons
from openlit.instrumentation.openai.async_openai import async_audio_create
-from openlit.instrumentation.openai.azure_openai import azure_chat_completions, azure_completions
-from openlit.instrumentation.openai.azure_openai import azure_image_generate, azure_embedding
-from openlit.instrumentation.openai.async_azure_openai import azure_async_chat_completions
-from openlit.instrumentation.openai.async_azure_openai import azure_async_completions
-from openlit.instrumentation.openai.async_azure_openai import azure_async_image_generate
-from openlit.instrumentation.openai.async_azure_openai import azure_async_embedding
-
_instruments = ("openai >= 1.1.1",)
class OpenAIInstrumentor(BaseInstrumentor):
@@ -40,7 +32,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.chat.completions",
"Completions.create",
- chat_wrapper(version, environment, application_name,
+ chat_completions(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
@@ -48,7 +40,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.chat.completions",
"AsyncCompletions.create",
- async_chat_wrapper(version, environment, application_name,
+ async_chat_completions(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
@@ -56,7 +48,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.images",
"Images.generate",
- image_generate_wrapper(version, environment, application_name,
+ image_generate(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
@@ -64,7 +56,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.images",
"AsyncImages.generate",
- async_image_generate_wrapper(version, environment, application_name,
+ async_image_generate(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
@@ -72,7 +64,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.embeddings",
"Embeddings.create",
- embedding_wrapper(version, environment, application_name,
+ embedding(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
@@ -80,24 +72,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.embeddings",
"AsyncEmbeddings.create",
- async_embedding_wrapper(version, environment, application_name,
- tracer, pricing_info, trace_content,
- metrics, disable_metrics),
- )
-
- wrap_function_wrapper(
- "openai.resources.completions",
- "Completions.create",
- azure_completions("azure_openai.completions", version, environment, application_name,
- tracer, pricing_info, trace_content,
- metrics, disable_metrics),
- )
-
- wrap_function_wrapper(
- "openai.resources.completions",
- "AsyncCompletions.create",
- azure_async_completions("azure_openai.completions", version,
- environment, application_name,
+ async_embedding(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
@@ -105,7 +80,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.images",
"Images.create_variation",
- image_variatons("openai.images.variations", version,
+ image_variatons(version,
environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
@@ -114,7 +89,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.images",
"AsyncImages.create_variation",
- async_image_variatons("openai.images.variations", version,
+ async_image_variatons(version,
environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
@@ -123,7 +98,7 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.audio.speech",
"Speech.create",
- audio_create("openai.audio.speech", version, environment, application_name,
+ audio_create(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
@@ -131,205 +106,11 @@ class OpenAIInstrumentor(BaseInstrumentor):
wrap_function_wrapper(
"openai.resources.audio.speech",
"AsyncSpeech.create",
- async_audio_create("openai.audio.speech", version, environment, application_name,
+ async_audio_create(version, environment, application_name,
tracer, pricing_info, trace_content,
metrics, disable_metrics),
)
- wrap_function_wrapper(
- "openai.resources.fine_tuning.jobs",
- "Jobs.create",
- finetune("openai.audio.speech", version, environment, application_name,
- tracer, pricing_info, trace_content,
- metrics, disable_metrics),
- )
-
- wrap_function_wrapper(
- "openai.resources.fine_tuning.jobs",
- "AsyncJobs.create",
- async_finetune("openai.fine_tuning.jo", version, environment, application_name,
- tracer, pricing_info, trace_content,
- metrics, disable_metrics),
- )
-
@staticmethod
def _uninstrument(self, **kwargs):
pass
-
-def chat_wrapper(version, environment, application_name, tracer, pricing_info, trace_content,
- metrics, disable_metrics):
- """
- Decorator for making a custom wrapper execute conditionally,
- based on whether the instance is for Azure OpenAI or not.
- """
- def wrapper(wrapped, instance, args, kwargs):
- # Default to using the standard OpenAI chat completions
- completion_func = chat_completions("openai.chat.completions", version, environment,
- application_name, tracer, pricing_info, trace_content,
- metrics, disable_metrics)
-
- # Check if it's an Azure instance by inspecting `base_url`
- try:
- base_url = getattr(instance, 'base_url', '')
- if 'azure.com' in base_url:
- # Switch to the Azure-specific chat completions logic
- completion_func = azure_chat_completions("azure_openai.chat.completions",
- version, environment, application_name,
- tracer, pricing_info, trace_content,
- metrics, disable_metrics)
- except AttributeError:
- pass # base_url attribute not found, proceed with the default
-
- # Execute the selected completion function
- return completion_func(wrapped, instance, args, kwargs)
-
- return wrapper
-
-def async_chat_wrapper(version, environment, application_name, tracer, pricing_info,
- trace_content, metrics, disable_metrics):
- """
- Decorator for making a custom wrapper execute conditionally,
- based on whether the instance is for Azure OpenAI or not.
- """
- def wrapper(wrapped, instance, args, kwargs):
- # Default to using the standard OpenAI chat completions
- completion_func = async_chat_completions("openai.chat.completions", version, environment,
- application_name, tracer,
- pricing_info, trace_content,
- metrics, disable_metrics)
-
- # Check if it's an Azure instance by inspecting `base_url`
- try:
- base_url = getattr(instance, 'base_url', '')
- if 'azure.com' in base_url:
- # Switch to the Azure-specific chat completions logic
- completion_func = azure_async_chat_completions("azure_openai.chat.completions",
- version, environment,
- application_name, tracer,
- pricing_info, trace_content,
- metrics, disable_metrics)
- except AttributeError:
- pass # base_url attribute not found, proceed with the default
-
- # Execute the selected completion function
- return completion_func(wrapped, instance, args, kwargs)
-
- return wrapper
-
-def image_generate_wrapper(version, environment, application_name, tracer, pricing_info,
- trace_content, metrics, disable_metrics):
- """
- Decorator for making a custom wrapper execute conditionally,
- based on whether the instance is for Azure OpenAI or not.
- """
- def wrapper(wrapped, instance, args, kwargs):
- # Default to using the standard OpenAI chat completions
- completion_func = image_generate("openai.images.generate", version, environment,
- application_name, tracer, pricing_info, trace_content,
- metrics, disable_metrics)
-
- # Check if it's an Azure instance by inspecting `base_url`
- try:
- base_url = getattr(instance, 'base_url', '')
- if 'azure.com' in base_url:
- # Switch to the Azure-specific chat completions logic
- completion_func = azure_image_generate("azure_openai.images.generate",
- version, environment, application_name,
- tracer, pricing_info, trace_content,
- metrics, disable_metrics)
- except AttributeError:
- pass # base_url attribute not found, proceed with the default
-
- # Execute the selected completion function
- return completion_func(wrapped, instance, args, kwargs)
-
- return wrapper
-
-def async_image_generate_wrapper(version, environment, application_name, tracer,
- pricing_info, trace_content, metrics, disable_metrics):
- """
- Decorator for making a custom wrapper execute conditionally,
- based on whether the instance is for Azure OpenAI or not.
- """
- def wrapper(wrapped, instance, args, kwargs):
- # Default to using the standard OpenAI chat completions
- completion_func = async_image_generate("openai.images.generate", version,
- environment, application_name, tracer,
- pricing_info, trace_content,
- metrics, disable_metrics)
-
- # Check if it's an Azure instance by inspecting `base_url`
- try:
- base_url = getattr(instance, 'base_url', '')
- if 'azure.com' in base_url:
- # Switch to the Azure-specific chat completions logic
- completion_func = azure_async_image_generate("azure_openai.images.generate",
- version, environment,
- application_name, tracer,
- pricing_info, trace_content,
- metrics, disable_metrics)
- except AttributeError:
- pass # base_url attribute not found, proceed with the default
-
- # Execute the selected completion function
- return completion_func(wrapped, instance, args, kwargs)
-
- return wrapper
-
-def embedding_wrapper(version, environment, application_name, tracer, pricing_info,
- trace_content, metrics, disable_metrics):
- """
- Decorator for making a custom wrapper execute conditionally,
- based on whether the instance is for Azure OpenAI or not.
- """
- def wrapper(wrapped, instance, args, kwargs):
- # Default to using the standard OpenAI chat completions
- completion_func = embedding("openai.embeddings", version, environment,
- application_name, tracer, pricing_info, trace_content,
- metrics, disable_metrics)
-
- # Check if it's an Azure instance by inspecting `base_url`
- try:
- base_url = getattr(instance, 'base_url', '')
- if 'azure.com' in base_url:
- # Switch to the Azure-specific chat completions logic
- completion_func = azure_embedding("azure_openai.embeddings",
- version, environment, application_name,
- tracer, pricing_info, trace_content,
- metrics, disable_metrics)
- except AttributeError:
- pass # base_url attribute not found, proceed with the default
-
- # Execute the selected completion function
- return completion_func(wrapped, instance, args, kwargs)
-
- return wrapper
-
-def async_embedding_wrapper(version, environment, application_name, tracer,
- pricing_info, trace_content, metrics, disable_metrics):
- """
- Decorator for making a custom wrapper execute conditionally,
- based on whether the instance is for Azure OpenAI or not.
- """
- def wrapper(wrapped, instance, args, kwargs):
- # Default to using the standard OpenAI chat completions
- completion_func = async_embedding("openai.embeddings", version, environment,
- application_name, tracer, pricing_info, trace_content,
- metrics, disable_metrics)
-
- # Check if it's an Azure instance by inspecting `base_url`
- try:
- base_url = getattr(instance, 'base_url', '')
- if 'azure.com' in base_url:
- # Switch to the Azure-specific chat completions logic
- completion_func = azure_async_embedding("azure_openai.embeddings", version,
- environment, application_name, tracer,
- pricing_info, trace_content,
- metrics, disable_metrics)
- except AttributeError:
- pass # base_url attribute not found, proceed with the default
-
- # Execute the selected completion function
- return completion_func(wrapped, instance, args, kwargs)
-
- return wrapper
diff --git a/sdk/python/src/openlit/instrumentation/openai/async_azure_openai.py b/sdk/python/src/openlit/instrumentation/openai/async_azure_openai.py
deleted file mode 100644
index 16704f98..00000000
--- a/sdk/python/src/openlit/instrumentation/openai/async_azure_openai.py
+++ /dev/null
@@ -1,900 +0,0 @@
-# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, too-many-branches
-"""
-Module for monitoring Azure OpenAI API calls.
-"""
-
-import logging
-from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
-from openlit.__helpers import get_chat_model_cost, get_embed_model_cost
-from openlit.__helpers import get_image_model_cost, openai_tokens, handle_exception
-from openlit.semcov import SemanticConvetion
-
-# Initialize logger for logging potential issues and operations
-logger = logging.getLogger(__name__)
-
-def azure_async_chat_completions(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for chat completions to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the chat completions method to add telemetry.
- """
-
- async def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'chat.completions' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and token usage, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'chat.completions' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'chat.completions' method.
- kwargs: Keyword arguments for the 'chat.completions' method.
-
- Returns:
- The response from the original 'chat.completions' method.
- """
-
- # Check if streaming is enabled for the API call
- streaming = kwargs.get("stream", False)
-
- # pylint: disable=no-else-return
- if streaming:
- # Special handling for streaming response to accommodate the nature of data flow
- async def stream_generator():
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- # Placeholder for aggregating streaming response
- llmresponse = ""
-
- # Loop through streaming events capturing relevant details
- async for chunk in await wrapped(*args, **kwargs):
- # Collect message IDs and aggregated response from events
- if len(chunk.choices) > 0:
- # pylint: disable=line-too-long
- if hasattr(chunk.choices[0], "delta") and hasattr(chunk.choices[0].delta, "content"):
- content = chunk.choices[0].delta.content
- if content:
- llmresponse += content
- yield chunk
- response_id = chunk.id
- model = "azure_" + chunk.model
-
- # Handling exception ensure observability without disrupting operation
- try:
- # Format 'messages' into a single string
- message_prompt = kwargs.get("messages", "")
- formatted_messages = []
- for message in message_prompt:
- role = message["role"]
- content = message["content"]
-
- if isinstance(content, list):
- content_str = ", ".join(
- # pylint: disable=line-too-long
- f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
- if "type" in item else f'text: {item["text"]}'
- for item in content
- )
- formatted_messages.append(f"{role}: {content_str}")
- else:
- formatted_messages.append(f"{role}: {content}")
- prompt = "\n".join(formatted_messages)
-
- # Calculate tokens using input prompt and aggregated response
- prompt_tokens = openai_tokens(prompt,
- kwargs.get("model", "gpt-3.5-turbo"))
- completion_tokens = openai_tokens(llmresponse,
- kwargs.get("model", "gpt-3.5-turbo"))
-
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- prompt_tokens, completion_tokens)
-
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response_id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- prompt_tokens + completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
- },
- )
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: llmresponse,
- },
- )
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(prompt_tokens + completion_tokens, attributes)
- metrics["genai_completion_tokens"].add(completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- return stream_generator()
-
- # Handling for non-streaming responses
- else:
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = await wrapped(*args, **kwargs)
-
- try:
- # Find base model from response
- model = "azure_" + response.model
-
- # Format 'messages' into a single string
- message_prompt = kwargs.get("messages", "")
- formatted_messages = []
- for message in message_prompt:
- role = message["role"]
- content = message["content"]
-
- if isinstance(content, list):
- content_str = ", ".join(
- # pylint: disable=line-too-long
- f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
- if "type" in item else f'text: {item["text"]}'
- for item in content
- )
- formatted_messages.append(f"{role}: {content_str}")
- else:
- formatted_messages.append(f"{role}: {content}")
- prompt = "\n".join(formatted_messages)
-
- # Set base span attribues
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- False)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
- },
- )
-
- # Set span attributes when tools is not passed to the function call
- if "tools" not in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- [response.choices[0].finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- # Set span attributes for when n = 1 (default)
- if "n" not in kwargs or kwargs["n"] == 1:
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[0].message.content,
- },
- )
-
- # Set span attributes for when n > 0
- else:
- i = 0
- while i < kwargs["n"] and trace_content is True:
- attribute_name = f"gen_ai.content.completion.{i}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[i].message.content,
- },
- )
- i += 1
-
- # Return original response
- return response
-
- # Set span attributes when tools is passed to the function call
- elif "tools" in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
- },
- )
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response.usage.total_tokens, attributes)
- metrics["genai_completion_tokens"].add(response.usage.completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(response.usage.prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def azure_async_completions(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for completions to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the chat completions method to add telemetry.
- """
-
- async def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'completions' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and token usage, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'completions' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'completions' method.
- kwargs: Keyword arguments for the 'completions' method.
-
- Returns:
- The response from the original 'chat.completions' method.
- """
-
- # Check if streaming is enabled for the API call
- streaming = kwargs.get("stream", False)
-
- # pylint: disable=no-else-return
- if streaming:
- # Special handling for streaming response to accommodate the nature of data flow
- async def stream_generator():
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- # Placeholder for aggregating streaming response
- llmresponse = ""
-
- # Loop through streaming events capturing relevant details
- async for chunk in await wrapped(*args, **kwargs):
- # Collect message IDs and aggregated response from events
- if len(chunk.choices) > 0:
- if hasattr(chunk.choices[0], "text"):
- content = chunk.choices[0].text
- if content:
- llmresponse += content
- yield chunk
- response_id = chunk.id
- model = "azure_" + chunk.model
-
- # Handling exception ensure observability without disrupting operation
- try:
- prompt = kwargs.get("prompt", "")
-
- # Calculate tokens using input prompt and aggregated response
- prompt_tokens = openai_tokens(prompt,
- "gpt-3.5-turbo")
- completion_tokens = openai_tokens(llmresponse,
- "gpt-3.5-turbo")
-
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- prompt_tokens, completion_tokens)
-
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response_id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- prompt_tokens + completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
- },
- )
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: llmresponse,
- },
- )
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(prompt_tokens + completion_tokens, attributes)
- metrics["genai_completion_tokens"].add(completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- return stream_generator()
-
- # Handling for non-streaming responses
- else:
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = wrapped(*args, **kwargs)
-
- try:
- # Find base model from response
- model = "azure_" + response.model
-
- # Set base span attribues
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- False)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
- },
- )
-
- # Set span attributes when tools is not passed to the function call
- if "tools" not in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- [response.choices[0].finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- # Set span attributes for when n = 1 (default)
- if "n" not in kwargs or kwargs["n"] == 1:
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[0].text,
- },
- )
-
- # Set span attributes for when n > 0
- else:
- i = 0
- while i < kwargs["n"] and trace_content is True:
- attribute_name = f"gen_ai.content.completion.{i}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[i].text,
- },
- )
- i += 1
- return response
-
- # Set span attributes when tools is passed to the function call
- elif "tools" in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
- },
- )
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response.usage.total_tokens, attributes)
- metrics["genai_completion_tokens"].add(response.usage.completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(response.usage.prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def azure_async_embedding(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for embeddings to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the embeddings method to add telemetry.
- """
-
- async def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'embeddings' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and token usage, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'embeddings' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'embeddings' method.
- kwargs: Keyword arguments for the 'embeddings' method.
-
- Returns:
- The response from the original 'embeddings' method.
- """
-
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = await wrapped(*args, **kwargs)
-
- try:
- # Calculate cost of the operation
- cost = get_embed_model_cost("azure_" + response.model,
- pricing_info, response.usage.prompt_tokens)
-
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- "azure_" + response.model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
- kwargs.get("encoding_format", "float"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
- kwargs.get("dimensions", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST, cost)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("input", ""),
- },
- )
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- "azure_" + response.model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response.usage.total_tokens, attributes)
- metrics["genai_prompt_tokens"].add(response.usage.prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def azure_async_image_generate(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for image generation to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI image generation.
- trace_content: Flag indicating whether to trace the input prompt and generated images.
-
- Returns:
- A function that wraps the image generation method to add telemetry.
- """
-
- async def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'images.generate' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'images.generate' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'images.generate' method.
- kwargs: Keyword arguments for the 'images.generate' method.
-
- Returns:
- The response from the original 'images.generate' method.
- """
-
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = await wrapped(*args, **kwargs)
- images_count = 0
-
- try:
- # Find Image format
- if "response_format" in kwargs and kwargs["response_format"] == "b64_json":
- image = "b64_json"
- else:
- image = "url"
-
- # Calculate cost of the operation
- cost = get_image_model_cost("azure_" + kwargs.get("model", "dall-e-3"),
- pricing_info, kwargs.get("size", "1024x1024"),
- kwargs.get("quality", "standard"))
-
- for items in response.data:
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.created)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- "azure_" + kwargs.get("model", "dall-e-3"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_SIZE,
- kwargs.get("size", "1024x1024"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
- kwargs.get("quality", "standard"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_STYLE,
- kwargs.get("style", "vivid"))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_REVISED_PROMPT,
- items.revised_prompt if response.revised_prompt else "")
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
- },
- )
- attribute_name = f"gen_ai.response.image.{images_count}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
- },
- )
-
- images_count+=1
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- len(response.data) * cost)
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- "azure_" + kwargs.get("model", "dall-e-3")
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
diff --git a/sdk/python/src/openlit/instrumentation/openai/async_openai.py b/sdk/python/src/openlit/instrumentation/openai/async_openai.py
index 0a959352..d5e28b6a 100644
--- a/sdk/python/src/openlit/instrumentation/openai/async_openai.py
+++ b/sdk/python/src/openlit/instrumentation/openai/async_openai.py
@@ -1,11 +1,13 @@
-# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, too-many-branches
"""
Module for monitoring OpenAI API calls.
"""
import logging
+import time
+from urllib.parse import urlparse
+from typing import Tuple, Any
from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
from openlit.__helpers import (
get_chat_model_cost,
get_embed_model_cost,
@@ -14,19 +16,47 @@ from openlit.__helpers import (
openai_tokens,
handle_exception,
response_as_dict,
+ calculate_ttft,
+ calculate_tbt,
+ create_metrics_attributes,
)
from openlit.semcov import SemanticConvetion
# Initialize logger for logging potential issues and operations
logger = logging.getLogger(__name__)
-def async_chat_completions(gen_ai_endpoint, version, environment, application_name,
+def set_server_address_and_port(client_instance: Any) -> Tuple[str, int]:
+ """
+ Determines and returns the server address and port based on the provided client's `base_url`,
+ using defaults if none found or values are None.
+ """
+ default_server_address = "api.openai.com"
+ default_server_port = 443
+
+ base_client = getattr(client_instance, "_client", None)
+ base_url = getattr(base_client, "base_url", None)
+
+ if base_url:
+ if isinstance(base_url, str):
+ url = urlparse(base_url)
+ server_address = url.hostname or default_server_address
+ server_port = url.port if url.port is not None else default_server_port
+ else: # base_url might not be a str; handle as an object.
+ server_address = getattr(base_url, "host", None) or default_server_address
+ port_attr = getattr(base_url, "port", None)
+ server_port = port_attr if port_attr is not None else default_server_port
+ else: # no base_url provided; use defaults.
+ server_address = default_server_address
+ server_port = default_server_port
+
+ return server_address, server_port
+
+def async_chat_completions(version, environment, application_name,
tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for chat completions to collect metrics.
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
@@ -54,6 +84,8 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
wrapped,
span,
kwargs,
+ server_address,
+ server_port,
**args,
):
self.__wrapped__ = wrapped
@@ -61,9 +93,20 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
# Placeholder for aggregating streaming response
self._llmresponse = ""
self._response_id = ""
+ self._response_model = ""
+ self._finish_reason = ""
+ self._openai_response_service_tier = ""
+ self._openai_system_fingerprint = ""
self._args = args
self._kwargs = kwargs
+ self._start_time = time.time()
+ self._end_time = None
+ self._timestamps = []
+ self._ttft = 0
+ self._tbt = 0
+ self._server_address = server_address
+ self._server_port = server_port
async def __aenter__(self):
await self.__wrapped__.__aenter__()
@@ -82,6 +125,14 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
async def __anext__(self):
try:
chunk = await self.__wrapped__.__anext__()
+ end_time = time.time()
+ # Record the timestamp for the current chunk
+ self._timestamps.append(end_time)
+
+ if len(self._timestamps) == 1:
+ # Calculate time to first chunk
+ self._ttft = calculate_ttft(self._timestamps, self._start_time)
+
chunked = response_as_dict(chunk)
# Collect message IDs and aggregated response from events
if (len(chunked.get('choices')) > 0 and ('delta' in chunked.get('choices')[0] and
@@ -91,10 +142,18 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
if content:
self._llmresponse += content
self._response_id = chunked.get('id')
+ self._response_model = chunked.get('model')
+ self._finish_reason = chunked.get('choices')[0].get('finish_reason')
+ self._openai_response_service_tier = chunked.get('service_tier')
+ self._openai_system_fingerprint = chunked.get('system_fingerprint')
return chunk
except StopAsyncIteration:
# Handling exception ensure observability without disrupting operation
try:
+ self._end_time = time.time()
+ if len(self._timestamps) > 1:
+ self._tbt = calculate_tbt(self._timestamps)
+
# Format 'messages' into a single string
message_prompt = self._kwargs.get("messages", "")
formatted_messages = []
@@ -103,68 +162,92 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
content = message["content"]
if isinstance(content, list):
- content_str = ", ".join(
- # pylint: disable=line-too-long
- f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
- if "type" in item else f'text: {item["text"]}'
- for item in content
- )
+ content_str_list = []
+ for item in content:
+ if item["type"] == "text":
+ content_str_list.append(f'text: {item["text"]}')
+ elif (item["type"] == "image_url" and
+ not item["image_url"]["url"].startswith("data:")):
+ content_str_list.append(f'image_url: {item["image_url"]["url"]}')
+ content_str = ", ".join(content_str_list)
formatted_messages.append(f"{role}: {content_str}")
else:
formatted_messages.append(f"{role}: {content}")
prompt = "\n".join(formatted_messages)
+ request_model = self._kwargs.get("model", "gpt-4o")
+
# Calculate tokens using input prompt and aggregated response
- prompt_tokens = openai_tokens(prompt,
- self._kwargs.get("model", "gpt-3.5-turbo"))
- completion_tokens = openai_tokens(self._llmresponse,
- self._kwargs.get("model", "gpt-3.5-turbo"))
+ input_tokens = openai_tokens(prompt,
+ request_model)
+ output_tokens = openai_tokens(self._llmresponse,
+ request_model)
# Calculate cost of the operation
- cost = get_chat_model_cost(self._kwargs.get("model", "gpt-3.5-turbo"),
- pricing_info, prompt_tokens,
- completion_tokens)
+ cost = get_chat_model_cost(request_model,
+ pricing_info, input_tokens,
+ output_tokens)
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
+ self._kwargs.get("seed", ""))
+ self._span.set_attribute(SemanticConvetion.SERVER_PORT,
+ self._server_port)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
+ self._kwargs.get("frequency_penalty", 0.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
+ self._kwargs.get("max_tokens", -1))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
+ self._kwargs.get("presence_penalty", 0.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_STOP_SEQUENCES,
+ self._kwargs.get("stop", []))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
+ self._kwargs.get("temperature", 1.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
+ self._kwargs.get("top_p", 1.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
+ [self._finish_reason])
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
self._response_id)
- self._span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ self._response_model)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+ input_tokens)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
+ output_tokens)
+ self._span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ self._server_address)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_REQUEST_SERVICE_TIER,
+ self._kwargs.get("service_tier", "auto"))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SERVICE_TIER,
+ self._openai_response_service_tier)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SYSTEM_FINGERPRINT,
+ self._openai_system_fingerprint)
+
+ # Set Span attributes (Extra)
+ self._span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- self._span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ self._span.set_attribute(SERVICE_NAME,
application_name)
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- self._kwargs.get("model", "gpt-3.5-turbo"))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
self._kwargs.get("user", ""))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
- self._kwargs.get("top_p", 1.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
- self._kwargs.get("max_tokens", -1))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- self._kwargs.get("temperature", 1.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- self._kwargs.get("presence_penalty", 0.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- self._kwargs.get("frequency_penalty", 0.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- self._kwargs.get("seed", ""))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- prompt_tokens + completion_tokens)
+ input_tokens + output_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TBT,
+ self._tbt)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TTFT,
+ self._ttft)
if trace_content:
self._span.add_event(
name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
@@ -178,31 +261,35 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: self._llmresponse,
},
)
-
self._span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- self._kwargs.get("model", "gpt-3.5-turbo")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=self._server_address,
+ server_port=self._server_port,
+ response_model=self._response_model,
+ )
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(
- prompt_tokens + completion_tokens, attributes
+ metrics["genai_client_usage_tokens"].record(
+ input_tokens + output_tokens, attributes
+ )
+ metrics["genai_client_operation_duration"].record(
+ self._end_time - self._start_time, attributes
+ )
+ metrics["genai_server_tbt"].record(
+ self._tbt, attributes
+ )
+ metrics["genai_server_ttft"].record(
+ self._ttft, attributes
)
- metrics["genai_completion_tokens"].add(completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(prompt_tokens, attributes)
+ metrics["genai_requests"].add(1, attributes)
+ metrics["genai_completion_tokens"].add(output_tokens, attributes)
+ metrics["genai_prompt_tokens"].add(input_tokens, attributes)
metrics["genai_cost"].record(cost, attributes)
except Exception as e:
@@ -231,20 +318,25 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
# Check if streaming is enabled for the API call
streaming = kwargs.get("stream", False)
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "gpt-4o")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
# pylint: disable=no-else-return
if streaming:
# Special handling for streaming response to accommodate the nature of data flow
awaited_wrapped = await wrapped(*args, **kwargs)
- span = tracer.start_span(gen_ai_endpoint, kind=SpanKind.CLIENT)
+ span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
- return TracedAsyncStream(awaited_wrapped, span, kwargs)
+ return TracedAsyncStream(awaited_wrapped, span, kwargs, server_address, server_port)
# Handling for non-streaming responses
else:
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = await wrapped(*args, **kwargs)
+ end_time = time.time()
response_dict = response_as_dict(response)
@@ -258,7 +350,6 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
if isinstance(content, list):
content_str = ", ".join(
- # pylint: disable=line-too-long
f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
if "type" in item else f'text: {item["text"]}'
for item in content
@@ -268,38 +359,70 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
formatted_messages.append(f"{role}: {content}")
prompt = "\n".join(formatted_messages)
- # Set base span attribues
+ input_tokens = response_dict.get('usage').get('prompt_tokens')
+ output_tokens = response_dict.get('usage').get('completion_tokens')
+
+ # Calculate cost of the operation
+ cost = get_chat_model_cost(request_model,
+ pricing_info, input_tokens,
+ output_tokens)
+
+ # Set base span attribues (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
+ kwargs.get("seed", ""))
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
+ kwargs.get("frequency_penalty", 0.0))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
+ kwargs.get("max_tokens", -1))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
+ kwargs.get("presence_penalty", 0.0))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_STOP_SEQUENCES,
+ kwargs.get("stop", []))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
+ kwargs.get("temperature", 1.0))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
+ kwargs.get("top_p", 1.0))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response_dict.get("id"))
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ response_dict.get('model'))
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+ input_tokens)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
+ output_tokens)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_REQUEST_SERVICE_TIER,
+ kwargs.get("service_tier", "auto"))
+ span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SERVICE_TIER,
+ response_dict.get('service_tier'))
+ span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SYSTEM_FINGERPRINT,
+ response_dict.get('system_fingerprint'))
+
+ # Set base span attribues (Extras)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "gpt-3.5-turbo"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
- kwargs.get("top_p", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
- kwargs.get("max_tokens", -1))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
+ input_tokens + output_tokens)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
+ cost)
+ span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TTFT,
+ end_time - start_time)
if trace_content:
span.add_event(
name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
@@ -308,93 +431,47 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
},
)
- # Set span attributes when tools is not passed to the function call
- if "tools" not in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(kwargs.get("model", "gpt-3.5-turbo"),
- pricing_info, response_dict.get('usage', {}).get('prompt_tokens', None),
- response_dict.get('usage', {}).get('completion_tokens', None))
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response_dict.get('usage', {}).get('total_tokens', None))
+ for i in range(kwargs.get('n',1)):
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- [response_dict.get('choices', [])[0].get('finish_reason', None)])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- # Set span attributes for when n = 1 (default)
- if "n" not in kwargs or kwargs["n"] == 1:
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response_dict.get('choices', [])[0].get("message").get("content"),
- },
- )
-
- # Set span attributes for when n > 0
- else:
- i = 0
- while i < kwargs["n"] and trace_content is True:
- attribute_name = f"gen_ai.content.completion.{i}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response_dict.get('choices')[i].get("message").get("content"),
- },
- )
- i += 1
-
- # Return original response
- return response
-
- # Set span attributes when tools is passed to the function call
- elif "tools" in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(kwargs.get("model", "gpt-3.5-turbo"),
- pricing_info, response_dict.get('usage').get('prompt_tokens'),
- response_dict.get('usage').get('completion_tokens'))
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
- },
- )
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response_dict.get('usage').get('completion_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response_dict.get('usage').get('total_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
+ [response_dict.get('choices')[i].get('finish_reason')])
+ if trace_content:
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: str(response_dict.get('choices')[i].get('message').get('content')),
+ },
+ )
+ if kwargs.get('tools'):
+ span.set_attribute(SemanticConvetion.GEN_AI_TOOL_CALLS,
+ str(response_dict.get('choices')[i].get('message').get('tool_calls')))
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "gpt-3.5-turbo")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=response_dict.get('model'),
+ )
+ metrics["genai_client_usage_tokens"].record(
+ input_tokens + output_tokens, attributes
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
+ metrics["genai_server_ttft"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response_dict.get('usage').get('total_tokens'), attributes)
- metrics["genai_completion_tokens"].add(response_dict.get('usage').get('completion_tokens'), attributes)
- metrics["genai_prompt_tokens"].add(response_dict.get('usage').get('prompt_tokens'), attributes)
+ metrics["genai_completion_tokens"].add(output_tokens, attributes)
+ metrics["genai_prompt_tokens"].add(input_tokens, attributes)
metrics["genai_cost"].record(cost, attributes)
# Return original response
@@ -409,20 +486,19 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
return wrapper
-def async_embedding(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
+def async_embedding(version, environment, application_name,
+ tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for embeddings to collect metrics.
-
+
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
tracer: OpenTelemetry tracer for creating spans.
pricing_info: Information used for calculating the cost of OpenAI usage.
trace_content: Flag indicating whether to trace the actual content.
-
+
Returns:
A function that wraps the embeddings method to add telemetry.
"""
@@ -444,38 +520,53 @@ def async_embedding(gen_ai_endpoint, version, environment, application_name,
The response from the original 'embeddings' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "text-embedding-ada-002")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = await wrapped(*args, **kwargs)
+ end_time = time.time()
+
response_dict = response_as_dict(response)
try:
+ request_model = kwargs.get("model", "text-embedding-ada-002")
+ input_tokens = response_dict.get('usage').get('prompt_tokens')
+
# Calculate cost of the operation
- cost = get_embed_model_cost(kwargs.get("model", "text-embedding-ada-002"),
- pricing_info, response_dict.get('usage').get('prompt_tokens'))
+ cost = get_embed_model_cost(request_model,
+ pricing_info, input_tokens)
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
+ kwargs.get('encoding_format', 'float'))
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ response_dict.get('model'))
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+ input_tokens)
+
+ # Set Span attributes (Extras)
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "text-embedding-ada-002"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
- kwargs.get("encoding_format", "float"))
- # span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
- # kwargs.get("dimensions", "null"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response_dict.get('usage').get('prompt_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response_dict.get('usage').get('total_tokens'))
+ input_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
@@ -490,26 +581,25 @@ def async_embedding(gen_ai_endpoint, version, environment, application_name,
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "text-embedding-ada-002")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=response_dict.get('model'),
+ )
+ metrics["genai_client_usage_tokens"].record(
+ input_tokens, attributes
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(
- response_dict.get('usage').get('total_tokens'), attributes)
- metrics["genai_prompt_tokens"].add(
- response_dict.get('usage').get('prompt_tokens'), attributes)
+ metrics["genai_prompt_tokens"].add(input_tokens, attributes)
metrics["genai_cost"].record(cost, attributes)
# Return original response
@@ -524,125 +614,19 @@ def async_embedding(gen_ai_endpoint, version, environment, application_name,
return wrapper
-def async_finetune(gen_ai_endpoint, version, environment, application_name,
+def async_image_generate(version, environment, application_name,
tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
- Generates a telemetry wrapper for fine-tuning jobs to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the fine tuning creation method to add telemetry.
- """
-
- async def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'fine_tuning.jobs.create' API call to add telemetry.
-
- This collects metrics such as execution time, usage stats, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'fine_tuning.jobs.create' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the method.
- kwargs: Keyword arguments for the method.
-
- Returns:
- The response from the original 'fine_tuning.jobs.create' method.
- """
-
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = await wrapped(*args, **kwargs)
-
- # Handling exception ensure observability without disrupting operation
- try:
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FINETUNING)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "gpt-3.5-turbo"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TRAINING_FILE,
- kwargs.get("training_file", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_VALIDATION_FILE,
- kwargs.get("validation_file", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_BATCH_SIZE,
- kwargs.get("hyperparameters.batch_size", "auto"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_MODEL_LRM,
- kwargs.get("hyperparameters.learning_rate_multiplier",
- "auto"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_MODEL_EPOCHS,
- kwargs.get("hyperparameters.n_epochs", "auto"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_MODEL_SUFFIX,
- kwargs.get("suffix", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_STATUS,
- response.status)
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_FINETUNING,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "gpt-3.5-turbo")
- }
-
- metrics["genai_requests"].add(1, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def async_image_generate(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
Generates a telemetry wrapper for image generation to collect metrics.
-
+
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
tracer: OpenTelemetry tracer for creating spans.
pricing_info: Information used for calculating the cost of OpenAI image generation.
trace_content: Flag indicating whether to trace the input prompt and generated images.
-
+
Returns:
A function that wraps the image generation method to add telemetry.
"""
@@ -664,8 +648,16 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
The response from the original 'images.generate' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "dall-e-2")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = await wrapped(*args, **kwargs)
+ end_time = time.time()
+
images_count = 0
try:
@@ -675,28 +667,38 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
else:
image = "url"
+ request_model = kwargs.get("model", "dall-e-2")
+
# Calculate cost of the operation
- cost = get_image_model_cost(kwargs.get("model", "dall-e-2"),
+ cost = get_image_model_cost(request_model,
pricing_info, kwargs.get("size", "1024x1024"),
kwargs.get("quality", "standard"))
for items in response.data:
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.created)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+ "image")
+
+ # Set Span attributes (Extras)
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "dall-e-2"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_SIZE,
kwargs.get("size", "1024x1024"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
@@ -707,6 +709,7 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
items.revised_prompt if items.revised_prompt else "")
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
+
if trace_content:
span.add_event(
name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
@@ -714,7 +717,7 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
},
)
- attribute_name = f"gen_ai.response.image.{images_count}"
+ attribute_name = f"{SemanticConvetion.GEN_AI_RESPONSE_IMAGE}.{images_count}"
span.add_event(
name=attribute_name,
attributes={
@@ -729,21 +732,20 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "dall-e-2")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=request_model,
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
metrics["genai_cost"].record(cost, attributes)
@@ -759,20 +761,19 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
return wrapper
-def async_image_variatons(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
+def async_image_variatons(version, environment, application_name,
+ tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for creating image variations to collect metrics.
-
+
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
tracer: OpenTelemetry tracer for creating spans.
pricing_info: Information used for calculating the cost of generating image variations.
trace_content: Flag indicating whether to trace the input image and generated variations.
-
+
Returns:
A function that wraps the image variations creation method to add telemetry.
"""
@@ -794,8 +795,16 @@ def async_image_variatons(gen_ai_endpoint, version, environment, application_nam
The response from the original 'images.create.variations' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "dall-e-2")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = await wrapped(*args, **kwargs)
+ end_time = time.time()
+
images_count = 0
try:
@@ -806,34 +815,43 @@ def async_image_variatons(gen_ai_endpoint, version, environment, application_nam
image = "url"
# Calculate cost of the operation
- cost = get_image_model_cost(kwargs.get("model", "dall-e-2"), pricing_info,
+ cost = get_image_model_cost(request_model, pricing_info,
kwargs.get("size", "1024x1024"), "standard")
for items in response.data:
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.created)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+ "image")
+
+ # Set Span attributes (Extras)
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "dall-e-2"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_SIZE,
kwargs.get("size", "1024x1024"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
"standard")
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
+ kwargs.get("user", ""))
+
if trace_content:
- attribute_name = f"gen_ai.response.image.{images_count}"
+ attribute_name = f"{SemanticConvetion.GEN_AI_RESPONSE_IMAGE}.{images_count}"
span.add_event(
name=attribute_name,
attributes={
@@ -848,21 +866,20 @@ def async_image_variatons(gen_ai_endpoint, version, environment, application_nam
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "dall-e-2")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=request_model,
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
metrics["genai_cost"].record(cost, attributes)
@@ -878,20 +895,19 @@ def async_image_variatons(gen_ai_endpoint, version, environment, application_nam
return wrapper
-def async_audio_create(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
+def async_audio_create(version, environment, application_name,
+ tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for creating speech audio to collect metrics.
-
+
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
tracer: OpenTelemetry tracer for creating spans.
pricing_info: Information used for calculating the cost of generating speech audio.
trace_content: Flag indicating whether to trace the input text and generated audio.
-
+
Returns:
A function that wraps the speech audio creation method to add telemetry.
"""
@@ -913,28 +929,42 @@ def async_audio_create(gen_ai_endpoint, version, environment, application_name,
The response from the original 'audio.speech.create' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "tts-1")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+ start_time = time.time()
response = await wrapped(*args, **kwargs)
+ end_time = time.time()
try:
# Calculate cost of the operation
- cost = get_audio_model_cost(kwargs.get("model", "tts-1"),
+ cost = get_audio_model_cost(request_model,
pricing_info, kwargs.get("input", ""))
# Set Span attributes
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AUDIO)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+ "speech")
+
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "tts-1"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_VOICE,
kwargs.get("voice", "alloy"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT,
@@ -954,21 +984,20 @@ def async_audio_create(gen_ai_endpoint, version, environment, application_name,
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_AUDIO,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "tts-1")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=request_model,
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
metrics["genai_cost"].record(cost, attributes)
diff --git a/sdk/python/src/openlit/instrumentation/openai/azure_openai.py b/sdk/python/src/openlit/instrumentation/openai/azure_openai.py
deleted file mode 100644
index eebd9b69..00000000
--- a/sdk/python/src/openlit/instrumentation/openai/azure_openai.py
+++ /dev/null
@@ -1,898 +0,0 @@
-# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, too-many-branches
-"""
-Module for monitoring Azure OpenAI API calls.
-"""
-
-import logging
-from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
-from openlit.__helpers import get_chat_model_cost, get_embed_model_cost
-from openlit.__helpers import get_image_model_cost, openai_tokens, handle_exception
-from openlit.semcov import SemanticConvetion
-
-# Initialize logger for logging potential issues and operations
-logger = logging.getLogger(__name__)
-
-def azure_chat_completions(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for chat completions to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the chat completions method to add telemetry.
- """
-
- def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'chat.completions' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and token usage, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'chat.completions' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'chat.completions' method.
- kwargs: Keyword arguments for the 'chat.completions' method.
-
- Returns:
- The response from the original 'chat.completions' method.
- """
-
- # Check if streaming is enabled for the API call
- streaming = kwargs.get("stream", False)
-
- # pylint: disable=no-else-return
- if streaming:
- # Special handling for streaming response to accommodate the nature of data flow
- def stream_generator():
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- # Placeholder for aggregating streaming response
- llmresponse = ""
-
- # Loop through streaming events capturing relevant details
- for chunk in wrapped(*args, **kwargs):
- # Collect message IDs and aggregated response from events
- if len(chunk.choices) > 0:
- # pylint: disable=line-too-long
- if hasattr(chunk.choices[0], "delta") and hasattr(chunk.choices[0].delta, "content"):
- content = chunk.choices[0].delta.content
- if content:
- llmresponse += content
- yield chunk
- response_id = chunk.id
- model = "azure_" + chunk.model
-
- # Handling exception ensure observability without disrupting operation
- try:
- # Format 'messages' into a single string
- message_prompt = kwargs.get("messages", "")
- formatted_messages = []
- for message in message_prompt:
- role = message["role"]
- content = message["content"]
-
- if isinstance(content, list):
- content_str = ", ".join(
- # pylint: disable=line-too-long
- f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
- if "type" in item else f'text: {item["text"]}'
- for item in content
- )
- formatted_messages.append(f"{role}: {content_str}")
- else:
- formatted_messages.append(f"{role}: {content}")
- prompt = "\n".join(formatted_messages)
-
- # Calculate tokens using input prompt and aggregated response
- prompt_tokens = openai_tokens(prompt,
- kwargs.get("model", "gpt-3.5-turbo"))
- completion_tokens = openai_tokens(llmresponse,
- kwargs.get("model", "gpt-3.5-turbo"))
-
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- prompt_tokens, completion_tokens)
-
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response_id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- prompt_tokens + completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
- },
- )
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: llmresponse,
- },
- )
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(prompt_tokens + completion_tokens, attributes)
- metrics["genai_completion_tokens"].add(completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- return stream_generator()
-
- # Handling for non-streaming responses
- else:
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = wrapped(*args, **kwargs)
-
- try:
- # Find base model from response
- model = "azure_" + response.model
-
- # Format 'messages' into a single string
- message_prompt = kwargs.get("messages", "")
- formatted_messages = []
- for message in message_prompt:
- role = message["role"]
- content = message["content"]
-
- if isinstance(content, list):
- content_str = ", ".join(
- # pylint: disable=line-too-long
- f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
- if "type" in item else f'text: {item["text"]}'
- for item in content
- )
- formatted_messages.append(f"{role}: {content_str}")
- else:
- formatted_messages.append(f"{role}: {content}")
- prompt = "\n".join(formatted_messages)
-
- # Set base span attribues
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- False)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
- },
- )
- if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
-
- # Set span attributes when tools is not passed to the function call
- if "tools" not in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- [response.choices[0].finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- # Set span attributes for when n = 1 (default)
- if "n" not in kwargs or kwargs["n"] == 1:
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[0].message.content,
- },
- )
-
- # Set span attributes for when n > 0
- else:
- i = 0
- while i < kwargs["n"] and trace_content is True:
- attribute_name = f"gen_ai.content.completion.{i}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[i].message.content,
- },
- )
- i += 1
-
- # Return original response
- return response
-
- # Set span attributes when tools is passed to the function call
- elif "tools" in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
- },
- )
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response.usage.total_tokens, attributes)
- metrics["genai_completion_tokens"].add(response.usage.completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(response.usage.prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def azure_completions(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for completions to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the chat completions method to add telemetry.
- """
-
- def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'completions' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and token usage, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'completions' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'completions' method.
- kwargs: Keyword arguments for the 'completions' method.
-
- Returns:
- The response from the original 'chat.completions' method.
- """
-
- # Check if streaming is enabled for the API call
- streaming = kwargs.get("stream", False)
-
- # pylint: disable=no-else-return
- if streaming:
- # Special handling for streaming response to accommodate the nature of data flow
- def stream_generator():
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- # Placeholder for aggregating streaming response
- llmresponse = ""
-
- # Loop through streaming events capturing relevant details
- for chunk in wrapped(*args, **kwargs):
- # Collect message IDs and aggregated response from events
- if len(chunk.choices) > 0:
- if hasattr(chunk.choices[0], "text"):
- content = chunk.choices[0].text
- if content:
- llmresponse += content
- yield chunk
- response_id = chunk.id
- model = "azure_" + chunk.model
-
- # Handling exception ensure observability without disrupting operation
- try:
- prompt = kwargs.get("prompt", "")
-
- # Calculate tokens using input prompt and aggregated response
- prompt_tokens = openai_tokens(prompt, "gpt-3.5-turbo")
- completion_tokens = openai_tokens(llmresponse, "gpt-3.5-turbo")
-
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- prompt_tokens, completion_tokens)
-
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response_id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- prompt_tokens + completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
- },
- )
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: llmresponse,
- },
- )
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(prompt_tokens + completion_tokens, attributes)
- metrics["genai_completion_tokens"].add(completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- return stream_generator()
-
- # Handling for non-streaming responses
- else:
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = wrapped(*args, **kwargs)
-
- try:
- # Find base model from response
- model = "azure_" + response.model
-
- # Set base span attribues
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOOL_CHOICE,
- kwargs.get("tool_choice", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
- False)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
- },
- )
-
- # Set span attributes when tools is not passed to the function call
- if "tools" not in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- [response.choices[0].finish_reason])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- # Set span attributes for when n = 1 (default)
- if "n" not in kwargs or kwargs["n"] == 1:
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[0].text,
- },
- )
-
- # Set span attributes for when n > 0
- else:
- i = 0
- while i < kwargs["n"] and trace_content is True:
- attribute_name = f"gen_ai.content.completion.{i}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[i].text,
- },
- )
- i += 1
- return response
-
- # Set span attributes when tools is passed to the function call
- elif "tools" in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(model, pricing_info,
- response.usage.prompt_tokens,
- response.usage.completion_tokens)
-
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- "Function called with tools")
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response.usage.total_tokens, attributes)
- metrics["genai_completion_tokens"].add(response.usage.completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(response.usage.prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def azure_embedding(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for embeddings to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the embeddings method to add telemetry.
- """
-
- def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'embeddings' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and token usage, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'embeddings' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'embeddings' method.
- kwargs: Keyword arguments for the 'embeddings' method.
-
- Returns:
- The response from the original 'embeddings' method.
- """
-
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = wrapped(*args, **kwargs)
-
- try:
- # Calculate cost of the operation
- cost = get_embed_model_cost("azure_" + response.model,
- pricing_info, response.usage.prompt_tokens)
-
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- "azure_" + response.model)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
- kwargs.get("encoding_format", "float"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
- kwargs.get("dimensions", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response.usage.total_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("input", ""),
- },
- )
-
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- "azure_" + response.model
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response.usage.total_tokens, attributes)
- metrics["genai_prompt_tokens"].add(response.usage.prompt_tokens, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def azure_image_generate(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for image generation to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI image generation.
- trace_content: Flag indicating whether to trace the input prompt and generated images.
-
- Returns:
- A function that wraps the image generation method to add telemetry.
- """
-
- def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'images.generate' API call to add telemetry.
-
- This collects metrics such as execution time, cost, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'images.generate' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the 'images.generate' method.
- kwargs: Keyword arguments for the 'images.generate' method.
-
- Returns:
- The response from the original 'images.generate' method.
- """
-
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = wrapped(*args, **kwargs)
- images_count = 0
-
- try:
- # Find Image format
- if "response_format" in kwargs and kwargs["response_format"] == "b64_json":
- image = "b64_json"
- else:
- image = "url"
-
- # Calculate cost of the operation
- cost = get_image_model_cost("azure_" + kwargs.get("model", "dall-e-3"),
- pricing_info, kwargs.get("size", "1024x1024"),
- kwargs.get("quality", "standard"))
-
- for items in response.data:
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.created)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- "azure_" + kwargs.get("model", "dall-e-3"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_SIZE,
- kwargs.get("size", "1024x1024"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
- kwargs.get("quality", "standard"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_STYLE,
- kwargs.get("style", "vivid"))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_REVISED_PROMPT,
- items.revised_prompt if response.revised_prompt else "")
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
- },
- )
- attribute_name = f"gen_ai.response.image.{images_count}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
- },
- )
-
- images_count+=1
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- len(response.data) * cost)
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_AZURE_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- "azure_" + kwargs.get("model", "dall-e-3")
- }
-
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_cost"].record(cost, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
diff --git a/sdk/python/src/openlit/instrumentation/openai/openai.py b/sdk/python/src/openlit/instrumentation/openai/openai.py
index 461d7415..105cd7b8 100644
--- a/sdk/python/src/openlit/instrumentation/openai/openai.py
+++ b/sdk/python/src/openlit/instrumentation/openai/openai.py
@@ -1,11 +1,13 @@
-# pylint: disable=duplicate-code, broad-exception-caught, too-many-statements, unused-argument, too-many-branches
"""
Module for monitoring OpenAI API calls.
"""
import logging
+import time
+from urllib.parse import urlparse
+from typing import Tuple, Any
from opentelemetry.trace import SpanKind, Status, StatusCode
-from opentelemetry.sdk.resources import TELEMETRY_SDK_NAME
+from opentelemetry.sdk.resources import SERVICE_NAME, TELEMETRY_SDK_NAME, DEPLOYMENT_ENVIRONMENT
from openlit.__helpers import (
get_chat_model_cost,
get_embed_model_cost,
@@ -14,19 +16,47 @@ from openlit.__helpers import (
openai_tokens,
handle_exception,
response_as_dict,
+ calculate_ttft,
+ calculate_tbt,
+ create_metrics_attributes,
)
from openlit.semcov import SemanticConvetion
# Initialize logger for logging potential issues and operations
logger = logging.getLogger(__name__)
-def chat_completions(gen_ai_endpoint, version, environment, application_name,
+def set_server_address_and_port(client_instance: Any) -> Tuple[str, int]:
+ """
+ Determines and returns the server address and port based on the provided client's `base_url`,
+ using defaults if none found or values are None.
+ """
+ default_server_address = "api.openai.com"
+ default_server_port = 443
+
+ base_client = getattr(client_instance, "_client", None)
+ base_url = getattr(base_client, "base_url", None)
+
+ if base_url:
+ if isinstance(base_url, str):
+ url = urlparse(base_url)
+ server_address = url.hostname or default_server_address
+ server_port = url.port if url.port is not None else default_server_port
+ else: # base_url might not be a str; handle as an object.
+ server_address = getattr(base_url, "host", None) or default_server_address
+ port_attr = getattr(base_url, "port", None)
+ server_port = port_attr if port_attr is not None else default_server_port
+ else: # no base_url provided; use defaults.
+ server_address = default_server_address
+ server_port = default_server_port
+
+ return server_address, server_port
+
+def chat_completions(version, environment, application_name,
tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for chat completions to collect metrics.
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
@@ -54,6 +84,8 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
wrapped,
span,
kwargs,
+ server_address,
+ server_port,
**args,
):
self.__wrapped__ = wrapped
@@ -61,9 +93,20 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
# Placeholder for aggregating streaming response
self._llmresponse = ""
self._response_id = ""
+ self._response_model = ""
+ self._finish_reason = ""
+ self._openai_response_service_tier = ""
+ self._openai_system_fingerprint = ""
self._args = args
self._kwargs = kwargs
+ self._start_time = time.time()
+ self._end_time = None
+ self._timestamps = []
+ self._ttft = 0
+ self._tbt = 0
+ self._server_address = server_address
+ self._server_port = server_port
def __enter__(self):
self.__wrapped__.__enter__()
@@ -82,6 +125,14 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
def __next__(self):
try:
chunk = self.__wrapped__.__next__()
+ end_time = time.time()
+ # Record the timestamp for the current chunk
+ self._timestamps.append(end_time)
+
+ if len(self._timestamps) == 1:
+ # Calculate time to first chunk
+ self._ttft = calculate_ttft(self._timestamps, self._start_time)
+
chunked = response_as_dict(chunk)
# Collect message IDs and aggregated response from events
if (len(chunked.get('choices')) > 0 and ('delta' in chunked.get('choices')[0] and
@@ -91,10 +142,18 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
if content:
self._llmresponse += content
self._response_id = chunked.get('id')
+ self._response_model = chunked.get('model')
+ self._finish_reason = chunked.get('choices')[0].get('finish_reason')
+ self._openai_response_service_tier = chunked.get('service_tier')
+ self._openai_system_fingerprint = chunked.get('system_fingerprint')
return chunk
except StopIteration:
# Handling exception ensure observability without disrupting operation
try:
+ self._end_time = time.time()
+ if len(self._timestamps) > 1:
+ self._tbt = calculate_tbt(self._timestamps)
+
# Format 'messages' into a single string
message_prompt = self._kwargs.get("messages", "")
formatted_messages = []
@@ -109,7 +168,6 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
content_str_list.append(f'text: {item["text"]}')
elif (item["type"] == "image_url" and
not item["image_url"]["url"].startswith("data:")):
- # pylint: disable=line-too-long
content_str_list.append(f'image_url: {item["image_url"]["url"]}')
content_str = ", ".join(content_str_list)
formatted_messages.append(f"{role}: {content_str}")
@@ -117,57 +175,79 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
formatted_messages.append(f"{role}: {content}")
prompt = "\n".join(formatted_messages)
+ request_model = self._kwargs.get("model", "gpt-4o")
+
# Calculate tokens using input prompt and aggregated response
- prompt_tokens = openai_tokens(prompt,
- self._kwargs.get("model", "gpt-3.5-turbo"))
- completion_tokens = openai_tokens(self._llmresponse,
- self._kwargs.get("model", "gpt-3.5-turbo"))
+ input_tokens = openai_tokens(prompt,
+ request_model)
+ output_tokens = openai_tokens(self._llmresponse,
+ request_model)
# Calculate cost of the operation
- cost = get_chat_model_cost(self._kwargs.get("model", "gpt-3.5-turbo"),
- pricing_info, prompt_tokens,
- completion_tokens)
+ cost = get_chat_model_cost(request_model,
+ pricing_info, input_tokens,
+ output_tokens)
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
+ self._kwargs.get("seed", ""))
+ self._span.set_attribute(SemanticConvetion.SERVER_PORT,
+ self._server_port)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
+ self._kwargs.get("frequency_penalty", 0.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
+ self._kwargs.get("max_tokens", -1))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
+ self._kwargs.get("presence_penalty", 0.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_STOP_SEQUENCES,
+ self._kwargs.get("stop", []))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
+ self._kwargs.get("temperature", 1.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
+ self._kwargs.get("top_p", 1.0))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
+ [self._finish_reason])
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
self._response_id)
- self._span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ self._response_model)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+ input_tokens)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
+ output_tokens)
+ self._span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ self._server_address)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_REQUEST_SERVICE_TIER,
+ self._kwargs.get("service_tier", "auto"))
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SERVICE_TIER,
+ self._openai_response_service_tier)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SYSTEM_FINGERPRINT,
+ self._openai_system_fingerprint)
+
+ # Set Span attributes (Extra)
+ self._span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- self._span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ self._span.set_attribute(SERVICE_NAME,
application_name)
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- self._kwargs.get("model", "gpt-3.5-turbo"))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
self._kwargs.get("user", ""))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
- self._kwargs.get("top_p", 1.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
- self._kwargs.get("max_tokens", -1))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- self._kwargs.get("temperature", 1.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- self._kwargs.get("presence_penalty", 0.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- self._kwargs.get("frequency_penalty", 0.0))
- self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- self._kwargs.get("seed", ""))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- prompt_tokens + completion_tokens)
+ input_tokens + output_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TBT,
+ self._tbt)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TTFT,
+ self._ttft)
if trace_content:
self._span.add_event(
name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
@@ -181,31 +261,35 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: self._llmresponse,
},
)
-
self._span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- self._kwargs.get("model", "gpt-3.5-turbo")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=self._server_address,
+ server_port=self._server_port,
+ response_model=self._response_model,
+ )
- metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(
- prompt_tokens + completion_tokens, attributes
+ metrics["genai_client_usage_tokens"].record(
+ input_tokens + output_tokens, attributes
+ )
+ metrics["genai_client_operation_duration"].record(
+ self._end_time - self._start_time, attributes
)
- metrics["genai_completion_tokens"].add(completion_tokens, attributes)
- metrics["genai_prompt_tokens"].add(prompt_tokens, attributes)
+ metrics["genai_server_tbt"].record(
+ self._tbt, attributes
+ )
+ metrics["genai_server_ttft"].record(
+ self._ttft, attributes
+ )
+ metrics["genai_requests"].add(1, attributes)
+ metrics["genai_completion_tokens"].add(output_tokens, attributes)
+ metrics["genai_prompt_tokens"].add(input_tokens, attributes)
metrics["genai_cost"].record(cost, attributes)
except Exception as e:
@@ -234,20 +318,25 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
# Check if streaming is enabled for the API call
streaming = kwargs.get("stream", False)
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "gpt-4o")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT} {request_model}"
# pylint: disable=no-else-return
if streaming:
# Special handling for streaming response to accommodate the nature of data flow
awaited_wrapped = wrapped(*args, **kwargs)
- span = tracer.start_span(gen_ai_endpoint, kind=SpanKind.CLIENT)
+ span = tracer.start_span(span_name, kind=SpanKind.CLIENT)
- return TracedSyncStream(awaited_wrapped, span, kwargs)
+ return TracedSyncStream(awaited_wrapped, span, kwargs, server_address, server_port)
# Handling for non-streaming responses
else:
- # pylint: disable=line-too-long
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = wrapped(*args, **kwargs)
+ end_time = time.time()
response_dict = response_as_dict(response)
@@ -261,7 +350,6 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
if isinstance(content, list):
content_str = ", ".join(
- # pylint: disable=line-too-long
f'{item["type"]}: {item["text"] if "text" in item else item["image_url"]}'
if "type" in item else f'text: {item["text"]}'
for item in content
@@ -271,38 +359,70 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
formatted_messages.append(f"{role}: {content}")
prompt = "\n".join(formatted_messages)
- # Set base span attribues
+ input_tokens = response_dict.get('usage').get('prompt_tokens')
+ output_tokens = response_dict.get('usage').get('completion_tokens')
+
+ # Calculate cost of the operation
+ cost = get_chat_model_cost(request_model,
+ pricing_info, input_tokens,
+ output_tokens)
+
+ # Set base span attribues (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
+ kwargs.get("seed", ""))
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
+ kwargs.get("frequency_penalty", 0.0))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
+ kwargs.get("max_tokens", -1))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
+ kwargs.get("presence_penalty", 0.0))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_STOP_SEQUENCES,
+ kwargs.get("stop", []))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
+ kwargs.get("temperature", 1.0))
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
+ kwargs.get("top_p", 1.0))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response_dict.get("id"))
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ response_dict.get('model'))
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+ input_tokens)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
+ output_tokens)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_REQUEST_SERVICE_TIER,
+ kwargs.get("service_tier", "auto"))
+ span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SERVICE_TIER,
+ response_dict.get('service_tier'))
+ span.set_attribute(SemanticConvetion.GEN_AI_OPENAI_RESPONSE_SYSTEM_FINGERPRINT,
+ response_dict.get('system_fingerprint'))
+
+ # Set base span attribues (Extras)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "gpt-3.5-turbo"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_P,
- kwargs.get("top_p", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
- kwargs.get("max_tokens", -1))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
- kwargs.get("temperature", 1.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_PRESENCE_PENALTY,
- kwargs.get("presence_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FREQUENCY_PENALTY,
- kwargs.get("frequency_penalty", 0.0))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_SEED,
- kwargs.get("seed", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
+ input_tokens + output_tokens)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
+ cost)
+ span.set_attribute(SemanticConvetion.GEN_AI_SERVER_TTFT,
+ end_time - start_time)
if trace_content:
span.add_event(
name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
@@ -311,93 +431,47 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
},
)
- # Set span attributes when tools is not passed to the function call
- if "tools" not in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(kwargs.get("model", "gpt-3.5-turbo"),
- pricing_info, response_dict.get('usage', {}).get('prompt_tokens', None),
- response_dict.get('usage', {}).get('completion_tokens', None))
-
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response_dict.get('usage', {}).get('total_tokens', None))
+ for i in range(kwargs.get('n',1)):
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- [response_dict.get('choices', [])[0].get('finish_reason', None)])
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
-
- # Set span attributes for when n = 1 (default)
- if "n" not in kwargs or kwargs["n"] == 1:
- if trace_content:
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response_dict.get('choices', [])[0].get("message").get("content"),
- },
- )
-
- # Set span attributes for when n > 0
- else:
- i = 0
- while i < kwargs["n"] and trace_content is True:
- attribute_name = f"gen_ai.content.completion.{i}"
- span.add_event(
- name=attribute_name,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response_dict.get('choices')[i].get("message").get("content"),
- },
- )
- i += 1
-
- # Return original response
- return response
-
- # Set span attributes when tools is passed to the function call
- elif "tools" in kwargs:
- # Calculate cost of the operation
- cost = get_chat_model_cost(kwargs.get("model", "gpt-3.5-turbo"),
- pricing_info, response_dict.get('usage').get('prompt_tokens'),
- response_dict.get('usage').get('completion_tokens'))
- span.add_event(
- name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
- attributes={
- SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
- },
- )
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
- response_dict.get('usage').get('completion_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response_dict.get('usage').get('total_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
- cost)
+ [response_dict.get('choices')[i].get('finish_reason')])
+ if trace_content:
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: str(response_dict.get('choices')[i].get('message').get('content')),
+ },
+ )
+ if kwargs.get('tools'):
+ span.set_attribute(SemanticConvetion.GEN_AI_TOOL_CALLS,
+ str(response_dict.get('choices')[i].get('message').get('tool_calls')))
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "gpt-3.5-turbo")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=response_dict.get('model'),
+ )
+ metrics["genai_client_usage_tokens"].record(
+ input_tokens + output_tokens, attributes
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
+ metrics["genai_server_ttft"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(response_dict.get('usage').get('total_tokens'), attributes)
- metrics["genai_completion_tokens"].add(response_dict.get('usage').get('completion_tokens'), attributes)
- metrics["genai_prompt_tokens"].add(response_dict.get('usage').get('prompt_tokens'), attributes)
+ metrics["genai_completion_tokens"].add(output_tokens, attributes)
+ metrics["genai_prompt_tokens"].add(input_tokens, attributes)
metrics["genai_cost"].record(cost, attributes)
# Return original response
@@ -412,13 +486,12 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
return wrapper
-def embedding(gen_ai_endpoint, version, environment, application_name,
+def embedding(version, environment, application_name,
tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for embeddings to collect metrics.
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
@@ -447,38 +520,53 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
The response from the original 'embeddings' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "text-embedding-ada-002")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = wrapped(*args, **kwargs)
+ end_time = time.time()
+
response_dict = response_as_dict(response)
try:
+ request_model = kwargs.get("model", "text-embedding-ada-002")
+ input_tokens = response_dict.get('usage').get('prompt_tokens')
+
# Calculate cost of the operation
- cost = get_embed_model_cost(kwargs.get("model", "text-embedding-ada-002"),
- pricing_info, response_dict.get('usage').get('prompt_tokens'))
+ cost = get_embed_model_cost(request_model,
+ pricing_info, input_tokens)
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
+ kwargs.get('encoding_format', 'float'))
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ response_dict.get('model'))
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
+ input_tokens)
+
+ # Set Span attributes (Extras)
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "text-embedding-ada-002"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
- kwargs.get("encoding_format", "float"))
- # span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
- # kwargs.get("dimensions", "null"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response_dict.get('usage').get('prompt_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- response_dict.get('usage').get('total_tokens'))
+ input_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
@@ -493,26 +581,25 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "text-embedding-ada-002")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=response_dict.get('model'),
+ )
+ metrics["genai_client_usage_tokens"].record(
+ input_tokens, attributes
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
- metrics["genai_total_tokens"].add(
- response_dict.get('usage').get('total_tokens'), attributes)
- metrics["genai_prompt_tokens"].add(
- response_dict.get('usage').get('prompt_tokens'), attributes)
+ metrics["genai_prompt_tokens"].add(input_tokens, attributes)
metrics["genai_cost"].record(cost, attributes)
# Return original response
@@ -527,118 +614,12 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
return wrapper
-def finetune(gen_ai_endpoint, version, environment, application_name,
- tracer, pricing_info, trace_content, metrics, disable_metrics):
- """
- Generates a telemetry wrapper for fine-tuning jobs to collect metrics.
-
- Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
- version: Version of the monitoring package.
- environment: Deployment environment (e.g., production, staging).
- application_name: Name of the application using the OpenAI API.
- tracer: OpenTelemetry tracer for creating spans.
- pricing_info: Information used for calculating the cost of OpenAI usage.
- trace_content: Flag indicating whether to trace the actual content.
-
- Returns:
- A function that wraps the fine tuning creation method to add telemetry.
- """
-
- def wrapper(wrapped, instance, args, kwargs):
- """
- Wraps the 'fine_tuning.jobs.create' API call to add telemetry.
-
- This collects metrics such as execution time, usage stats, and handles errors
- gracefully, adding details to the trace for observability.
-
- Args:
- wrapped: The original 'fine_tuning.jobs.create' method to be wrapped.
- instance: The instance of the class where the original method is defined.
- args: Positional arguments for the method.
- kwargs: Keyword arguments for the method.
-
- Returns:
- The response from the original 'fine_tuning.jobs.create' method.
- """
-
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
- response = wrapped(*args, **kwargs)
-
- # Handling exception ensure observability without disrupting operation
- try:
- # Set Span attributes
- span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
- span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_FINETUNING)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
- environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
- application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "gpt-3.5-turbo"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TRAINING_FILE,
- kwargs.get("training_file", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_VALIDATION_FILE,
- kwargs.get("validation_file", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_BATCH_SIZE,
- kwargs.get("hyperparameters.batch_size", "auto"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_MODEL_LRM,
- kwargs.get("hyperparameters.learning_rate_multiplier",
- "auto"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_MODEL_EPOCHS,
- kwargs.get("hyperparameters.n_epochs", "auto"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_MODEL_SUFFIX,
- kwargs.get("suffix", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
- response.id)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- response.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_FINETUNE_STATUS,
- response.status)
- span.set_status(Status(StatusCode.OK))
-
- if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_FINETUNING,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "gpt-3.5-turbo")
- }
-
- metrics["genai_requests"].add(1, attributes)
-
- # Return original response
- return response
-
- except Exception as e:
- handle_exception(span, e)
- logger.error("Error in trace creation: %s", e)
-
- # Return original response
- return response
-
- return wrapper
-
-def image_generate(gen_ai_endpoint, version, environment, application_name,
+def image_generate(version, environment, application_name,
tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for image generation to collect metrics.
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
@@ -667,8 +648,16 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
The response from the original 'images.generate' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "dall-e-2")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = wrapped(*args, **kwargs)
+ end_time = time.time()
+
images_count = 0
try:
@@ -678,28 +667,38 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
else:
image = "url"
+ request_model = kwargs.get("model", "dall-e-2")
+
# Calculate cost of the operation
- cost = get_image_model_cost(kwargs.get("model", "dall-e-2"),
+ cost = get_image_model_cost(request_model,
pricing_info, kwargs.get("size", "1024x1024"),
kwargs.get("quality", "standard"))
for items in response.data:
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.created)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+ "image")
+
+ # Set Span attributes (Extras)
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "dall-e-2"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_SIZE,
kwargs.get("size", "1024x1024"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
@@ -710,6 +709,7 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
items.revised_prompt if items.revised_prompt else "")
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
+
if trace_content:
span.add_event(
name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
@@ -717,7 +717,7 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
},
)
- attribute_name = f"gen_ai.response.image.{images_count}"
+ attribute_name = f"{SemanticConvetion.GEN_AI_RESPONSE_IMAGE}.{images_count}"
span.add_event(
name=attribute_name,
attributes={
@@ -732,21 +732,20 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "dall-e-2")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=request_model,
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
metrics["genai_cost"].record(cost, attributes)
@@ -762,13 +761,12 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
return wrapper
-def image_variatons(gen_ai_endpoint, version, environment, application_name,
+def image_variatons(version, environment, application_name,
tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for creating image variations to collect metrics.
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
@@ -797,8 +795,16 @@ def image_variatons(gen_ai_endpoint, version, environment, application_name,
The response from the original 'images.create.variations' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "dall-e-2")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind= SpanKind.CLIENT) as span:
+ start_time = time.time()
response = wrapped(*args, **kwargs)
+ end_time = time.time()
+
images_count = 0
try:
@@ -809,34 +815,43 @@ def image_variatons(gen_ai_endpoint, version, environment, application_name,
image = "url"
# Calculate cost of the operation
- cost = get_image_model_cost(kwargs.get("model", "dall-e-2"), pricing_info,
+ cost = get_image_model_cost(request_model, pricing_info,
kwargs.get("size", "1024x1024"), "standard")
for items in response.data:
- # Set Span attributes
+ # Set Span attributes (OTel Semconv)
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.created)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+ "image")
+
+ # Set Span attributes (Extras)
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "dall-e-2"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
- kwargs.get("user", ""))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_SIZE,
kwargs.get("size", "1024x1024"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
"standard")
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
+ kwargs.get("user", ""))
+
if trace_content:
- attribute_name = f"gen_ai.response.image.{images_count}"
+ attribute_name = f"{SemanticConvetion.GEN_AI_RESPONSE_IMAGE}.{images_count}"
span.add_event(
name=attribute_name,
attributes={
@@ -851,21 +866,20 @@ def image_variatons(gen_ai_endpoint, version, environment, application_name,
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "dall-e-2")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=request_model,
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
metrics["genai_cost"].record(cost, attributes)
@@ -881,13 +895,12 @@ def image_variatons(gen_ai_endpoint, version, environment, application_name,
return wrapper
-def audio_create(gen_ai_endpoint, version, environment, application_name,
+def audio_create(version, environment, application_name,
tracer, pricing_info, trace_content, metrics, disable_metrics):
"""
Generates a telemetry wrapper for creating speech audio to collect metrics.
Args:
- gen_ai_endpoint: Endpoint identifier for logging and tracing.
version: Version of the monitoring package.
environment: Deployment environment (e.g., production, staging).
application_name: Name of the application using the OpenAI API.
@@ -916,28 +929,42 @@ def audio_create(gen_ai_endpoint, version, environment, application_name,
The response from the original 'audio.speech.create' method.
"""
- with tracer.start_as_current_span(gen_ai_endpoint, kind= SpanKind.CLIENT) as span:
+ server_address, server_port = set_server_address_and_port(instance)
+ request_model = kwargs.get("model", "tts-1")
+
+ span_name = f"{SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO} {request_model}"
+
+ with tracer.start_as_current_span(span_name, kind=SpanKind.CLIENT) as span:
+ start_time = time.time()
response = wrapped(*args, **kwargs)
+ end_time = time.time()
try:
# Calculate cost of the operation
- cost = get_audio_model_cost(kwargs.get("model", "tts-1"),
+ cost = get_audio_model_cost(request_model,
pricing_info, kwargs.get("input", ""))
# Set Span attributes
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO)
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_OPENAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AUDIO)
- span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
- gen_ai_endpoint)
- span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.SERVER_ADDRESS,
+ server_address)
+ span.set_attribute(SemanticConvetion.SERVER_PORT,
+ server_port)
+ span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_MODEL,
+ request_model)
+ span.set_attribute(SemanticConvetion.GEN_AI_OUTPUT_TYPE,
+ "speech")
+
+ span.set_attribute(DEPLOYMENT_ENVIRONMENT,
environment)
- span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
+ span.set_attribute(SERVICE_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
- kwargs.get("model", "tts-1"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_VOICE,
kwargs.get("voice", "alloy"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT,
@@ -957,21 +984,20 @@ def audio_create(gen_ai_endpoint, version, environment, application_name,
span.set_status(Status(StatusCode.OK))
if disable_metrics is False:
- attributes = {
- TELEMETRY_SDK_NAME:
- "openlit",
- SemanticConvetion.GEN_AI_APPLICATION_NAME:
- application_name,
- SemanticConvetion.GEN_AI_SYSTEM:
- SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
- SemanticConvetion.GEN_AI_ENVIRONMENT:
- environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_AUDIO,
- SemanticConvetion.GEN_AI_REQUEST_MODEL:
- kwargs.get("model", "tts-1")
- }
+ attributes = create_metrics_attributes(
+ service_name=application_name,
+ deployment_environment=environment,
+ operation=SemanticConvetion.GEN_AI_OPERATION_TYPE_AUDIO,
+ system=SemanticConvetion.GEN_AI_SYSTEM_OPENAI,
+ request_model=request_model,
+ server_address=server_address,
+ server_port=server_port,
+ response_model=request_model,
+ )
+ metrics["genai_client_operation_duration"].record(
+ end_time - start_time, attributes
+ )
metrics["genai_requests"].add(1, attributes)
metrics["genai_cost"].record(cost, attributes)
diff --git a/sdk/python/src/openlit/instrumentation/phidata/phidata.py b/sdk/python/src/openlit/instrumentation/phidata/phidata.py
index d444161e..2f97ccde 100644
--- a/sdk/python/src/openlit/instrumentation/phidata/phidata.py
+++ b/sdk/python/src/openlit/instrumentation/phidata/phidata.py
@@ -58,8 +58,8 @@ def phidata_wrap(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_PHIDATA)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_AGENT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_AGENT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
diff --git a/sdk/python/src/openlit/instrumentation/pinecone/pinecone.py b/sdk/python/src/openlit/instrumentation/pinecone/pinecone.py
index 47e41511..763cd53b 100644
--- a/sdk/python/src/openlit/instrumentation/pinecone/pinecone.py
+++ b/sdk/python/src/openlit/instrumentation/pinecone/pinecone.py
@@ -76,8 +76,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_VECTORDB)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
span.set_attribute(SemanticConvetion.DB_SYSTEM,
SemanticConvetion.DB_SYSTEM_PINECONE)
@@ -152,8 +152,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.DB_SYSTEM_PINECONE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_VECTORDB,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,
SemanticConvetion.DB_OPERATION:
db_operation
}
diff --git a/sdk/python/src/openlit/instrumentation/premai/premai.py b/sdk/python/src/openlit/instrumentation/premai/premai.py
index c7539114..33805b5b 100644
--- a/sdk/python/src/openlit/instrumentation/premai/premai.py
+++ b/sdk/python/src/openlit/instrumentation/premai/premai.py
@@ -113,8 +113,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_PREMAI)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -141,9 +141,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
self._kwargs.get("seed", ""))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -175,8 +175,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_PREMAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
self._kwargs.get("model", "gpt-3.5-turbo")
}
@@ -255,8 +255,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_PREMAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -298,9 +298,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
pricing_info, response_dict.usage.prompt_tokens,
response_dict.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.usage.total_tokens)
@@ -347,9 +347,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.usage.prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.usage.completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.usage.total_tokens)
@@ -368,8 +368,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_PREMAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "gpt-3.5-turbo")
}
@@ -439,8 +439,8 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_PREMAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -449,13 +449,13 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
application_name)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
kwargs.get("model", "text-embedding-3-large"))
- span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_FORMAT,
+ span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_ENCODING_FORMATS,
kwargs.get("encoding_format", "float"))
# span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_EMBEDDING_DIMENSION,
# kwargs.get("dimensions", "null"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.usage.prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.usage.total_tokens)
@@ -481,8 +481,8 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_PREMAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "text-embedding-ada-002")
}
diff --git a/sdk/python/src/openlit/instrumentation/qdrant/async_qdrant.py b/sdk/python/src/openlit/instrumentation/qdrant/async_qdrant.py
index 8e2c86cf..a4bd99eb 100644
--- a/sdk/python/src/openlit/instrumentation/qdrant/async_qdrant.py
+++ b/sdk/python/src/openlit/instrumentation/qdrant/async_qdrant.py
@@ -77,8 +77,8 @@ def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_VECTORDB)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
span.set_attribute(SemanticConvetion.DB_SYSTEM,
SemanticConvetion.DB_SYSTEM_QDRANT)
@@ -247,8 +247,8 @@ def async_general_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.DB_SYSTEM_QDRANT,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_VECTORDB,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,
SemanticConvetion.DB_OPERATION:
db_operation
}
diff --git a/sdk/python/src/openlit/instrumentation/qdrant/qdrant.py b/sdk/python/src/openlit/instrumentation/qdrant/qdrant.py
index a4d1669e..5dbb08bc 100644
--- a/sdk/python/src/openlit/instrumentation/qdrant/qdrant.py
+++ b/sdk/python/src/openlit/instrumentation/qdrant/qdrant.py
@@ -77,8 +77,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_VECTORDB)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB)
span.set_attribute(SemanticConvetion.DB_SYSTEM,
SemanticConvetion.DB_SYSTEM_QDRANT)
@@ -254,8 +254,8 @@ def general_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.DB_SYSTEM_QDRANT,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_VECTORDB,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_VECTORDB,
SemanticConvetion.DB_OPERATION:
db_operation
}
diff --git a/sdk/python/src/openlit/instrumentation/reka/async_reka.py b/sdk/python/src/openlit/instrumentation/reka/async_reka.py
index 28e195c9..e9215110 100644
--- a/sdk/python/src/openlit/instrumentation/reka/async_reka.py
+++ b/sdk/python/src/openlit/instrumentation/reka/async_reka.py
@@ -77,8 +77,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_REKAAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -111,9 +111,9 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(kwargs.get("model", "reka-core"),
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -134,8 +134,8 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_REKAAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "reka-core")
}
diff --git a/sdk/python/src/openlit/instrumentation/reka/reka.py b/sdk/python/src/openlit/instrumentation/reka/reka.py
index d5415bb9..f530a79f 100644
--- a/sdk/python/src/openlit/instrumentation/reka/reka.py
+++ b/sdk/python/src/openlit/instrumentation/reka/reka.py
@@ -77,8 +77,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_REKAAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -111,9 +111,9 @@ def chat(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(kwargs.get("model", "reka-core"),
pricing_info, prompt_tokens, completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -134,8 +134,8 @@ def chat(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_REKAAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "reka-core")
}
diff --git a/sdk/python/src/openlit/instrumentation/together/async_together.py b/sdk/python/src/openlit/instrumentation/together/async_together.py
index 6c083f53..a7eeb9bc 100644
--- a/sdk/python/src/openlit/instrumentation/together/async_together.py
+++ b/sdk/python/src/openlit/instrumentation/together/async_together.py
@@ -129,8 +129,8 @@ def async_completion(gen_ai_endpoint, version, environment, application_name,
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -160,9 +160,9 @@ def async_completion(gen_ai_endpoint, version, environment, application_name,
self._kwargs.get("seed", ""))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
self._prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
self._completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
self._total_tokens)
@@ -194,8 +194,8 @@ def async_completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
self._kwargs.get("model",
"meta-llama/Llama-3.3-70B-Instruct-Turbo")
@@ -280,8 +280,8 @@ def async_completion(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -328,9 +328,9 @@ def async_completion(gen_ai_endpoint, version, environment, application_name,
response_dict.get('usage', {}).get('prompt_tokens', None),
response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage', {}).get('completion_tokens', None))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage', {}).get('total_tokens', None))
@@ -380,9 +380,9 @@ def async_completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage').get('completion_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -401,8 +401,8 @@ def async_completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "meta-llama/Llama-3.3-70B-Instruct-Turbo")
}
@@ -489,8 +489,8 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -536,8 +536,8 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "black-forest-labs/FLUX.1-dev")
}
diff --git a/sdk/python/src/openlit/instrumentation/together/together.py b/sdk/python/src/openlit/instrumentation/together/together.py
index 0f9d4625..08ebf9d3 100644
--- a/sdk/python/src/openlit/instrumentation/together/together.py
+++ b/sdk/python/src/openlit/instrumentation/together/together.py
@@ -129,8 +129,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
self._span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
self._span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER)
- self._span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ self._span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
self._span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
self._span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -160,9 +160,9 @@ def completion(gen_ai_endpoint, version, environment, application_name,
self._kwargs.get("seed", ""))
self._span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
self._prompt_tokens)
- self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
self._completion_tokens)
self._span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
self._total_tokens)
@@ -194,8 +194,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
self._kwargs.get("model",
"meta-llama/Llama-3.3-70B-Instruct-Turbo")
@@ -280,8 +280,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -328,9 +328,9 @@ def completion(gen_ai_endpoint, version, environment, application_name,
response_dict.get('usage', {}).get('prompt_tokens', None),
response_dict.get('usage', {}).get('completion_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage', {}).get('prompt_tokens', None))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage', {}).get('completion_tokens', None))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage', {}).get('total_tokens', None))
@@ -380,9 +380,9 @@ def completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_COMPLETION: "Function called with tools",
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response_dict.get('usage').get('prompt_tokens'))
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response_dict.get('usage').get('completion_tokens'))
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response_dict.get('usage').get('total_tokens'))
@@ -401,8 +401,8 @@ def completion(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "meta-llama/Llama-3.3-70B-Instruct-Turbo")
}
@@ -489,8 +489,8 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_IMAGE)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
@@ -536,8 +536,8 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_TOGETHER,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_IMAGE,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_IMAGE,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
kwargs.get("model", "black-forest-labs/FLUX.1-dev")
}
diff --git a/sdk/python/src/openlit/instrumentation/transformers/transformers.py b/sdk/python/src/openlit/instrumentation/transformers/transformers.py
index fd372cad..f7f20d13 100644
--- a/sdk/python/src/openlit/instrumentation/transformers/transformers.py
+++ b/sdk/python/src/openlit/instrumentation/transformers/transformers.py
@@ -78,8 +78,8 @@ def text_wrap(gen_ai_endpoint, version, environment, application_name,
environment)
span.set_attribute(SemanticConvetion.GEN_AI_APPLICATION_NAME,
application_name)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MODEL,
instance.model.config.name_or_path)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TEMPERATURE,
@@ -95,7 +95,7 @@ def text_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
},
)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
i = 0
@@ -122,7 +122,7 @@ def text_wrap(gen_ai_endpoint, version, environment, application_name,
completion_tokens += general_tokens(llm_response)
i=i+1
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens + completion_tokens)
@@ -138,8 +138,8 @@ def text_wrap(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_HUGGING_FACE,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
instance.model.config.name_or_path
}
diff --git a/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py b/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py
index b1d0f651..c1e63b25 100644
--- a/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py
+++ b/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py
@@ -82,8 +82,8 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -94,9 +94,9 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -128,8 +128,8 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -168,8 +168,8 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -180,9 +180,9 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage_metadata.prompt_token_count)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage_metadata.candidates_token_count)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage_metadata.total_token_count)
@@ -215,8 +215,8 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -312,8 +312,8 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -324,9 +324,9 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -358,8 +358,8 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -398,8 +398,8 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -410,9 +410,9 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage_metadata.prompt_token_count)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage_metadata.candidates_token_count)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage_metadata.total_token_count)
@@ -445,8 +445,8 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -528,8 +528,8 @@ def predict_async(gen_ai_endpoint, version, environment, application_name, trace
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -540,9 +540,9 @@ def predict_async(gen_ai_endpoint, version, environment, application_name, trace
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -574,8 +574,8 @@ def predict_async(gen_ai_endpoint, version, environment, application_name, trace
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -668,8 +668,8 @@ def predict_streaming_async(gen_ai_endpoint, version, environment, application_n
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -680,9 +680,9 @@ def predict_streaming_async(gen_ai_endpoint, version, environment, application_n
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -714,8 +714,8 @@ def predict_streaming_async(gen_ai_endpoint, version, environment, application_n
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -793,8 +793,8 @@ def start_chat_async(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -805,9 +805,9 @@ def start_chat_async(gen_ai_endpoint, version, environment, application_name, tr
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -839,8 +839,8 @@ def start_chat_async(gen_ai_endpoint, version, environment, application_name, tr
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -931,8 +931,8 @@ def start_chat_streaming_async(gen_ai_endpoint, version, environment, applicatio
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -943,9 +943,9 @@ def start_chat_streaming_async(gen_ai_endpoint, version, environment, applicatio
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -977,8 +977,8 @@ def start_chat_streaming_async(gen_ai_endpoint, version, environment, applicatio
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -1052,8 +1052,8 @@ def embeddings_async(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -1064,7 +1064,7 @@ def embeddings_async(gen_ai_endpoint, version, environment, application_name, tr
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens)
@@ -1090,8 +1090,8 @@ def embeddings_async(gen_ai_endpoint, version, environment, application_name, tr
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py b/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py
index c0da193f..75ee47e8 100644
--- a/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py
+++ b/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py
@@ -82,8 +82,8 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -94,9 +94,9 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -128,8 +128,8 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -168,8 +168,8 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -180,9 +180,9 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage_metadata.prompt_token_count)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage_metadata.candidates_token_count)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage_metadata.total_token_count)
@@ -215,8 +215,8 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -312,8 +312,8 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -324,9 +324,9 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -358,8 +358,8 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -398,8 +398,8 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -410,9 +410,9 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
response.usage_metadata.prompt_token_count)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
response.usage_metadata.candidates_token_count)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
response.usage_metadata.total_token_count)
@@ -445,8 +445,8 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -528,8 +528,8 @@ def predict(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -540,9 +540,9 @@ def predict(gen_ai_endpoint, version, environment, application_name, tracer,
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -574,8 +574,8 @@ def predict(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -668,8 +668,8 @@ def predict_streaming(gen_ai_endpoint, version, environment, application_name, t
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -680,9 +680,9 @@ def predict_streaming(gen_ai_endpoint, version, environment, application_name, t
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -714,8 +714,8 @@ def predict_streaming(gen_ai_endpoint, version, environment, application_name, t
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -793,8 +793,8 @@ def start_chat(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -805,9 +805,9 @@ def start_chat(gen_ai_endpoint, version, environment, application_name, tracer,
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -839,8 +839,8 @@ def start_chat(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -931,8 +931,8 @@ def start_chat_streaming(gen_ai_endpoint, version, environment, application_name
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -943,9 +943,9 @@ def start_chat_streaming(gen_ai_endpoint, version, environment, application_name
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
completion_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -977,8 +977,8 @@ def start_chat_streaming(gen_ai_endpoint, version, environment, application_name
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
@@ -1052,8 +1052,8 @@ def embeddings(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -1064,7 +1064,7 @@ def embeddings(gen_ai_endpoint, version, environment, application_name, tracer,
model)
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
prompt_tokens)
@@ -1090,8 +1090,8 @@ def embeddings(gen_ai_endpoint, version, environment, application_name, tracer,
SemanticConvetion.GEN_AI_SYSTEM_VERTEXAI,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_EMBEDDING,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_EMBEDDING,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/instrumentation/vllm/vllm.py b/sdk/python/src/openlit/instrumentation/vllm/vllm.py
index 835c32be..76d50e62 100644
--- a/sdk/python/src/openlit/instrumentation/vllm/vllm.py
+++ b/sdk/python/src/openlit/instrumentation/vllm/vllm.py
@@ -57,8 +57,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(TELEMETRY_SDK_NAME, "openlit")
span.set_attribute(SemanticConvetion.GEN_AI_SYSTEM,
SemanticConvetion.GEN_AI_SYSTEM_VLLM)
- span.set_attribute(SemanticConvetion.GEN_AI_TYPE,
- SemanticConvetion.GEN_AI_TYPE_CHAT)
+ span.set_attribute(SemanticConvetion.GEN_AI_OPERATION,
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT)
span.set_attribute(SemanticConvetion.GEN_AI_ENDPOINT,
gen_ai_endpoint)
span.set_attribute(SemanticConvetion.GEN_AI_ENVIRONMENT,
@@ -97,9 +97,9 @@ def generate(gen_ai_endpoint, version, environment, application_name,
total_tokens = input_tokens + output_tokens
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
input_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
@@ -118,8 +118,8 @@ def generate(gen_ai_endpoint, version, environment, application_name,
SemanticConvetion.GEN_AI_SYSTEM_VLLM,
SemanticConvetion.GEN_AI_ENVIRONMENT:
environment,
- SemanticConvetion.GEN_AI_TYPE:
- SemanticConvetion.GEN_AI_TYPE_CHAT,
+ SemanticConvetion.GEN_AI_OPERATION:
+ SemanticConvetion.GEN_AI_OPERATION_TYPE_CHAT,
SemanticConvetion.GEN_AI_REQUEST_MODEL:
model
}
diff --git a/sdk/python/src/openlit/otel/metrics.py b/sdk/python/src/openlit/otel/metrics.py
index 0a4d94d0..09bf5b60 100644
--- a/sdk/python/src/openlit/otel/metrics.py
+++ b/sdk/python/src/openlit/otel/metrics.py
@@ -18,6 +18,75 @@ else:
# Global flag to check if the meter provider initialization is complete.
METER_SET = False
+_GEN_AI_CLIENT_OPERATION_DURATION_BUCKETS = [
+ 0.01,
+ 0.02,
+ 0.04,
+ 0.08,
+ 0.16,
+ 0.32,
+ 0.64,
+ 1.28,
+ 2.56,
+ 5.12,
+ 10.24,
+ 20.48,
+ 40.96,
+ 81.92,
+]
+
+_GEN_AI_SERVER_TBT = [
+ 0.01,
+ 0.025,
+ 0.05,
+ 0.075,
+ 0.1,
+ 0.15,
+ 0.2,
+ 0.3,
+ 0.4,
+ 0.5,
+ 0.75,
+ 1.0,
+ 2.5
+]
+
+_GEN_AI_SERVER_TFTT = [
+ 0.001,
+ 0.005,
+ 0.01,
+ 0.02,
+ 0.04,
+ 0.06,
+ 0.08,
+ 0.1,
+ 0.25,
+ 0.5,
+ 0.75,
+ 1.0,
+ 2.5,
+ 5.0,
+ 7.5,
+ 10.0
+]
+
+_GEN_AI_CLIENT_TOKEN_USAGE_BUCKETS = [
+ 1,
+ 4,
+ 16,
+ 64,
+ 256,
+ 1024,
+ 4096,
+ 16384,
+ 65536,
+ 262144,
+ 1048576,
+ 4194304,
+ 16777216,
+ 67108864,
+]
+
def setup_meter(application_name, environment, meter, otlp_endpoint, otlp_headers):
"""
Sets up OpenTelemetry metrics with a counter for total requests.
@@ -73,26 +142,48 @@ def setup_meter(application_name, environment, meter, otlp_endpoint, otlp_header
# Define and create the metrics
metrics_dict = {
+ # OTel Semconv
+ "genai_client_usage_tokens": meter.create_histogram(
+ name=SemanticConvetion.GEN_AI_CLIENT_TOKEN_USAGE,
+ description="Measures number of input and output tokens used",
+ unit="{token}",
+ explicit_bucket_boundaries_advisory=_GEN_AI_CLIENT_TOKEN_USAGE_BUCKETS,
+ ),
+ "genai_client_operation_duration": meter.create_histogram(
+ name=SemanticConvetion.GEN_AI_CLIENT_OPERATION_DURATION,
+ description="GenAI operation duration",
+ unit="s",
+ explicit_bucket_boundaries_advisory=_GEN_AI_CLIENT_OPERATION_DURATION_BUCKETS,
+ ),
+ "genai_server_tbt": meter.create_histogram(
+ name=SemanticConvetion.GEN_AI_SERVER_TBT,
+ description="Time per output token generated after the first token for successful responses",
+ unit="s",
+ explicit_bucket_boundaries_advisory=_GEN_AI_SERVER_TBT,
+ ),
+ "genai_server_ttft": meter.create_histogram(
+ name=SemanticConvetion.GEN_AI_SERVER_TTFT,
+ description="Time to generate first token for successful responses",
+ unit="s",
+ explicit_bucket_boundaries_advisory=_GEN_AI_SERVER_TFTT,
+ ),
+
+ # Extra
"genai_requests": meter.create_counter(
name=SemanticConvetion.GEN_AI_REQUESTS,
description="Number of requests to GenAI",
unit="1",
),
"genai_prompt_tokens": meter.create_counter(
- name=SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ name=SemanticConvetion.GEN_AI_USAGE_INPUT_TOKENS,
description="Number of prompt tokens processed.",
unit="1",
),
"genai_completion_tokens": meter.create_counter(
- name=SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
+ name=SemanticConvetion.GEN_AI_USAGE_OUTPUT_TOKENS,
description="Number of completion tokens processed.",
unit="1",
),
- "genai_total_tokens": meter.create_counter(
- name=SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
- description="Number of total tokens processed.",
- unit="1",
- ),
"genai_cost": meter.create_histogram(
name=SemanticConvetion.GEN_AI_USAGE_COST,
description="The distribution of GenAI request costs.",
diff --git a/sdk/python/src/openlit/semcov/__init__.py b/sdk/python/src/openlit/semcov/__init__.py
index af599026..1e8bd696 100644
--- a/sdk/python/src/openlit/semcov/__init__.py
+++ b/sdk/python/src/openlit/semcov/__init__.py
@@ -16,98 +16,84 @@ class SemanticConvetion:
the application's data logging and processing functionalities.
"""
- # GenAI General
- GEN_AI_ENDPOINT = "gen_ai.endpoint"
- GEN_AI_SYSTEM = "gen_ai.system"
- GEN_AI_ENVIRONMENT = "gen_ai.environment"
- GEN_AI_APPLICATION_NAME = "gen_ai.application_name"
- GEN_AI_TYPE = "gen_ai.operation.name"
- GEN_AI_HUB_OWNER = "gen_ai.hub.owner"
- GEN_AI_HUB_REPO = "gen_ai.hub.repo"
- GEN_AI_RETRIEVAL_SOURCE = "gen_ai.retrieval.source"
- GEN_AI_REQUESTS = "gen_ai.total.requests"
- GEN_AI_DATA_SOURCES = "gen_ai.data_source_count"
+ # General Attributes (OTel Semconv)
+ SERVER_PORT = "server.port"
+ SERVER_ADDRESS = "server.address"
+ ERROR_TYPE = "error.type"
- # GenAI Request
+ # GenAI Metric Names (OTel Semconv)
+ GEN_AI_CLIENT_TOKEN_USAGE = "gen_ai.client.token.usage"
+ GEN_AI_CLIENT_OPERATION_DURATION = "gen_ai.client.operation.duration"
+ GEN_AI_SERVER_REQUEST_DURATION = "gen_ai.server.request.duration"
+ GEN_AI_SERVER_TBT = "gen_ai.server.time_per_output_token"
+ GEN_AI_SERVER_TTFT = "gen_ai.server.time_to_first_token"
+
+ # GenAI Request Attributes (OTel Semconv)
+ GEN_AI_OPERATION = "gen_ai.operation.name"
+ GEN_AI_SYSTEM = "gen_ai.system"
+ GEN_AI_OUTPUT_TYPE = "gen_ai.output.type"
+ GEN_AI_ENDPOINT = "gen_ai.endpoint"
GEN_AI_REQUEST_MODEL = "gen_ai.request.model"
- GEN_AI_REQUEST_TEMPERATURE = "gen_ai.request.temperature"
- GEN_AI_REQUEST_TOP_P = "gen_ai.request.top_p"
- GEN_AI_REQUEST_TOP_K = "gen_ai.request.top_k"
- GEN_AI_REQUEST_MAX_TOKENS = "gen_ai.request.max_tokens"
- GEN_AI_REQUEST_IS_STREAM = "gen_ai.request.is_stream"
- GEN_AI_REQUEST_USER = "gen_ai.request.user"
GEN_AI_REQUEST_SEED = "gen_ai.request.seed"
+ GEN_AI_REQUEST_ENCODING_FORMATS = "gen_ai.request.encoding_formats"
GEN_AI_REQUEST_FREQUENCY_PENALTY = "gen_ai.request.frequency_penalty"
+ GEN_AI_REQUEST_MAX_TOKENS = "gen_ai.request.max_tokens"
GEN_AI_REQUEST_PRESENCE_PENALTY = "gen_ai.request.presence_penalty"
- GEN_AI_REQUEST_EMBEDDING_FORMAT = "gen_ai.request.embedding_format"
- GEN_AI_REQUEST_EMBEDDING_DIMENSION = "gen_ai.request.embedding_dimension"
- GEN_AI_REQUEST_TOOL_CHOICE = "gen_ai.request.tool_choice"
- GEN_AI_REQUEST_AUDIO_VOICE = "gen_ai.request.audio_voice"
- GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT = "gen_ai.request.audio_response_format"
- GEN_AI_REQUEST_AUDIO_SPEED = "gen_ai.request.audio_speed"
- GEN_AI_REQUEST_AUDIO_SETTINGS = "gen_ai.request.audio_settings"
- GEN_AI_REQUEST_AUDIO_DURATION = "gen_ai.request.audio_duration"
- GEN_AI_REQUEST_FINETUNE_STATUS = "gen_ai.request.fine_tune_status"
- GEN_AI_REQUEST_FINETUNE_MODEL_SUFFIX = "gen_ai.request.fine_tune_model_suffix"
- GEN_AI_REQUEST_FINETUNE_MODEL_EPOCHS = "gen_ai.request.fine_tune_n_epochs"
- GEN_AI_REQUEST_FINETUNE_MODEL_LRM = "gen_ai.request.learning_rate_multiplier"
- GEN_AI_REQUEST_FINETUNE_BATCH_SIZE = "gen_ai.request.fine_tune_batch_size"
- GEN_AI_REQUEST_VALIDATION_FILE = "gen_ai.request.validation_file"
- GEN_AI_REQUEST_TRAINING_FILE = "gen_ai.request.training_file"
-
- GEN_AI_REQUEST_IMAGE_SIZE = "gen_ai.request.image_size"
- GEN_AI_REQUEST_IMAGE_QUALITY = "gen_ai.request.image_quality"
- GEN_AI_REQUEST_IMAGE_STYLE = "gen_ai.request.image_style"
-
- # GenAI Usage
- GEN_AI_USAGE_PROMPT_TOKENS = "gen_ai.usage.input_tokens"
- GEN_AI_USAGE_COMPLETION_TOKENS = "gen_ai.usage.output_tokens"
- GEN_AI_USAGE_TOTAL_TOKENS = "gen_ai.usage.total_tokens"
- GEN_AI_USAGE_COST = "gen_ai.usage.cost"
+ GEN_AI_REQUEST_STOP_SEQUENCES = "gen_ai.request.stop_sequences"
+ GEN_AI_REQUEST_TEMPERATURE = "gen_ai.request.temperature"
+ GEN_AI_REQUEST_TOP_K = "gen_ai.request.top_k"
+ GEN_AI_REQUEST_TOP_P = "gen_ai.request.top_p"
- # GenAI Response
- GEN_AI_RESPONSE_ID = "gen_ai.response.id"
+ # GenAI Response Attributes (OTel Semconv)
+ GEN_AI_TOKEN_TYPE = "gen_ai.token.type"
GEN_AI_RESPONSE_FINISH_REASON = "gen_ai.response.finish_reasons"
- GEN_AI_RESPONSE_IMAGE = "gen_ai.response.image" # Not used directly in code yet
-
- # GenAI Content
- GEN_AI_CONTENT_PROMPT_EVENT = "gen_ai.content.prompt"
- GEN_AI_CONTENT_PROMPT = "gen_ai.prompt"
- GEN_AI_CONTENT_COMPLETION_EVENT = "gen_ai.content.completion"
- GEN_AI_CONTENT_COMPLETION = "gen_ai.completion"
- GEN_AI_CONTENT_REVISED_PROMPT = "gen_ai.content.revised_prompt"
-
- # GenAI Rag
- GEN_AI_RAG_MAX_SEGMENTS = "gen_ai.rag.max_segments"
- GEN_AI_RAG_STRATEGY = "gen_ai.rag.strategy"
- GEN_AI_RAG_SIMILARITY_THRESHOLD = "gen_ai.rag.similarity_threshold"
- GEN_AI_RAG_MAX_NEIGHBORS = "gen_ai.rag.max_neighbors"
- GEN_AI_RAG_DOCUMENTS_PATH = "gen_ai.rag.documents_path"
- GEN_AI_RAG_FILE_IDS = "gen_ai.rag.file_ids"
+ GEN_AI_RESPONSE_ID = "gen_ai.response.id"
+ GEN_AI_RESPONSE_MODEL = "gen_ai.response.model"
+ GEN_AI_USAGE_INPUT_TOKENS = "gen_ai.usage.input_tokens"
+ GEN_AI_USAGE_OUTPUT_TOKENS = "gen_ai.usage.output_tokens"
+ GEN_AI_TOOL_CALL_ID = "gen_ai.tool.call.id"
+ GEN_AI_TOOL_NAME = "gen_ai.tool.name"
- # GenAI Evaluation Metrics
- GEN_AI_EVAL_CONTEXT_RELEVANCY = "gen_ai.eval.context_relevancy"
- GEN_AI_EVAL_ANSWER_RELEVANCY = "gen_ai.eval.answer_relevancy"
- GEN_AI_EVAL_GROUNDEDNESS = "gen_ai.eval.groundedness"
+ # GenAI Operation Types (OTel Semconv)
+ GEN_AI_OPERATION_TYPE_CHAT = "chat"
+ GEN_AI_OPERATION_TYPE_TOOLS = "execute_tool"
+ GEN_AI_OPERATION_TYPE_EMBEDDING = "embeddings"
+ GEN_AI_OPERATION_TYPE_IMAGE = "image"
+ GEN_AI_OPERATION_TYPE_AUDIO = "audio"
+ GEN_AI_OPERATION_TYPE_VECTORDB = "vectordb"
+ GEN_AI_OPERATION_TYPE_FRAMEWORK = "framework"
+ GEN_AI_OPERATION_TYPE_AGENT = "agent"
+ GEN_AI_OPERATION_TYPE_CREATE_AGENT = "create_agent"
- GEN_AI_TYPE_CHAT = "chat"
- GEN_AI_TYPE_EMBEDDING = "embedding"
- GEN_AI_TYPE_IMAGE = "image"
- GEN_AI_TYPE_AUDIO = "audio"
- GEN_AI_TYPE_FINETUNING = "fine_tuning"
- GEN_AI_TYPE_VECTORDB = "vectordb"
- GEN_AI_TYPE_FRAMEWORK = "framework"
- GEN_AI_TYPE_AGENT = "agent"
+ # GenAI Output Types (OTel Semconv)
+ GEN_AI_OUTPUT_TYPE_IMAGE = "image"
+ GEN_AI_OUTPUT_TYPE_JSON = "json"
+ GEN_AI_OUTPUT_TYPE_SPEECH = "speech"
+ GEN_AI_OUTPUT_TYPE_TEXT = "text"
- GEN_AI_SYSTEM_HUGGING_FACE = "huggingface"
- GEN_AI_SYSTEM_OPENAI = "openai"
- GEN_AI_SYSTEM_AZURE_OPENAI = "azure_openai"
+ # GenAI System Names (OTel Semconv)
GEN_AI_SYSTEM_ANTHROPIC = "anthropic"
+ GEN_AI_SYSTEM_AWS_BEDROCK = "aws.bedrock"
+ GEN_AI_SYSTEM_AZURE_AI_INFERENCE = "az.ai.inference"
+ GEN_AI_SYSTEM_AZURE_OPENAI = "az.ai.openai"
GEN_AI_SYSTEM_COHERE = "cohere"
- GEN_AI_SYSTEM_MISTRAL = "mistral"
- GEN_AI_SYSTEM_BEDROCK = "bedrock"
- GEN_AI_SYSTEM_VERTEXAI = "vertex_ai"
+ GEN_AI_SYSTEM_DEEPSEEK = "deepseek"
+ GEN_AI_SYSTEM_GEMINI = "gemini"
GEN_AI_SYSTEM_GROQ = "groq"
+ GEN_AI_SYSTEM_IBM_WATSON = "ibm.watson.ai"
+ GEN_AI_SYSTEM_MISTRAL = "mistral_ai"
+ GEN_AI_SYSTEM_OPENAI = "openai"
+ GEN_AI_SYSTEM_PERPLEXITY = "perplexity"
+ GEN_AI_SYSTEM_VERTEXAI = "vertex_ai"
+ GEN_AI_SYSTEM_XAI = "xai"
+
+ # GenAI OpenAI Attributes (OTel Semconv)
+ GEN_AI_OPENAI_REQUEST_SERVICE_TIER = "gen_ai.openai.request.service_tier"
+ GEN_AI_OPENAI_RESPONSE_SERVICE_TIER = "gen_ai.openai.response.service_tier"
+ GEN_AI_OPENAI_RESPONSE_SYSTEM_FINGERPRINT = "gen_ai.openai.response.system_fingerprint"
+
+ # GenAI System Names (Extra)
+ GEN_AI_SYSTEM_HUGGING_FACE = "huggingface"
GEN_AI_SYSTEM_OLLAMA = "ollama"
GEN_AI_SYSTEM_GPT4ALL = "gpt4all"
GEN_AI_SYSTEM_ELEVENLABS = "elevenlabs"
@@ -115,7 +101,6 @@ class SemanticConvetion:
GEN_AI_SYSTEM_GOOGLE_AI_STUDIO = "google-ai-studio"
GEN_AI_SYSTEM_REKAAI = "rekaai"
GEN_AI_SYSTEM_PREMAI = "premai"
- GEN_AI_SYSTEM_AZURE_AI_INFERENCE = "azure-ai-inference"
GEN_AI_SYSTEM_LANGCHAIN = "langchain"
GEN_AI_SYSTEM_LLAMAINDEX = "llama_index"
GEN_AI_SYSTEM_HAYSTACK = "haystack"
@@ -136,6 +121,53 @@ class SemanticConvetion:
GEN_AI_SYSTEM_LETTA = "letta"
GEN_AI_SYSTEM_TOGETHER = "together"
+ # GenAI Request Attributes (Extra)
+ GEN_AI_REQUEST_IS_STREAM = "gen_ai.request.is_stream"
+ GEN_AI_REQUEST_USER = "gen_ai.request.user"
+ GEN_AI_REQUEST_EMBEDDING_DIMENSION = "gen_ai.request.embedding_dimension"
+ GEN_AI_REQUEST_TOOL_CHOICE = "gen_ai.request.tool_choice"
+ GEN_AI_REQUEST_AUDIO_VOICE = "gen_ai.request.audio_voice"
+ GEN_AI_REQUEST_AUDIO_RESPONSE_FORMAT = "gen_ai.request.audio_response_format"
+ GEN_AI_REQUEST_AUDIO_SPEED = "gen_ai.request.audio_speed"
+ GEN_AI_REQUEST_AUDIO_SETTINGS = "gen_ai.request.audio_settings"
+ GEN_AI_REQUEST_AUDIO_DURATION = "gen_ai.request.audio_duration"
+ GEN_AI_REQUEST_IMAGE_SIZE = "gen_ai.request.image_size"
+ GEN_AI_REQUEST_IMAGE_QUALITY = "gen_ai.request.image_quality"
+ GEN_AI_REQUEST_IMAGE_STYLE = "gen_ai.request.image_style"
+ GEN_AI_HUB_OWNER = "gen_ai.hub.owner"
+ GEN_AI_HUB_REPO = "gen_ai.hub.repo"
+ GEN_AI_RETRIEVAL_SOURCE = "gen_ai.retrieval.source"
+ GEN_AI_REQUESTS = "gen_ai.total.requests"
+ GEN_AI_DATA_SOURCES = "gen_ai.data_source_count"
+ GEN_AI_ENVIRONMENT = "gen_ai.environment"
+ GEN_AI_APPLICATION_NAME = "gen_ai.application_name"
+
+ # GenAI Response Attributes (Extra)
+ GEN_AI_USAGE_TOTAL_TOKENS = "gen_ai.usage.total_tokens"
+ GEN_AI_USAGE_COST = "gen_ai.usage.cost"
+ GEN_AI_RESPONSE_IMAGE = "gen_ai.response.image"
+ GEN_AI_TOOL_CALLS = "gen_ai.response.tool_calls"
+
+ # GenAI Content
+ GEN_AI_CONTENT_PROMPT_EVENT = "gen_ai.content.prompt"
+ GEN_AI_CONTENT_PROMPT = "gen_ai.prompt"
+ GEN_AI_CONTENT_COMPLETION_EVENT = "gen_ai.content.completion"
+ GEN_AI_CONTENT_COMPLETION = "gen_ai.completion"
+ GEN_AI_CONTENT_REVISED_PROMPT = "gen_ai.content.revised_prompt"
+
+ # GenAI Rag
+ GEN_AI_RAG_MAX_SEGMENTS = "gen_ai.rag.max_segments"
+ GEN_AI_RAG_STRATEGY = "gen_ai.rag.strategy"
+ GEN_AI_RAG_SIMILARITY_THRESHOLD = "gen_ai.rag.similarity_threshold"
+ GEN_AI_RAG_MAX_NEIGHBORS = "gen_ai.rag.max_neighbors"
+ GEN_AI_RAG_DOCUMENTS_PATH = "gen_ai.rag.documents_path"
+ GEN_AI_RAG_FILE_IDS = "gen_ai.rag.file_ids"
+
+ # GenAI Evaluation Metrics
+ GEN_AI_EVAL_CONTEXT_RELEVANCY = "gen_ai.eval.context_relevancy"
+ GEN_AI_EVAL_ANSWER_RELEVANCY = "gen_ai.eval.answer_relevancy"
+ GEN_AI_EVAL_GROUNDEDNESS = "gen_ai.eval.groundedness"
+
# Vector DB
DB_OPERATION_API_ENDPOINT = "db.operation.api_endpoint"
DB_REQUESTS = "db.total.requests"
diff --git a/sdk/python/tests/test_ai21.py b/sdk/python/tests/test_ai21.py.hold
similarity index 100%
rename from sdk/python/tests/test_ai21.py
rename to sdk/python/tests/test_ai21.py.hold
diff --git a/sdk/python/tests/test_mem0.py b/sdk/python/tests/test_mem0.py.hold
similarity index 100%
rename from sdk/python/tests/test_mem0.py
rename to sdk/python/tests/test_mem0.py.hold
diff --git a/sdk/typescript/src/instrumentation/anthropic/wrapper.ts b/sdk/typescript/src/instrumentation/anthropic/wrapper.ts
index 18e9275c..120f65af 100644
--- a/sdk/typescript/src/instrumentation/anthropic/wrapper.ts
+++ b/sdk/typescript/src/instrumentation/anthropic/wrapper.ts
@@ -194,7 +194,7 @@ export default class AnthropicWrapper extends BaseWrapper {
}
const prompt = formattedMessages.join('\n');
- span.setAttribute(SemanticConvention.GEN_AI_TYPE, SemanticConvention.GEN_AI_TYPE_CHAT);
+ span.setAttribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT);
span.setAttribute(SemanticConvention.GEN_AI_RESPONSE_ID, result.id);
const model = result.model || 'claude-3-sonnet-20240229';
@@ -231,9 +231,9 @@ export default class AnthropicWrapper extends BaseWrapper {
}
// Request Params attributes : End
- span.setAttribute(SemanticConvention.GEN_AI_USAGE_PROMPT_TOKENS, result.usage.input_tokens);
+ span.setAttribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, result.usage.input_tokens);
span.setAttribute(
- SemanticConvention.GEN_AI_USAGE_COMPLETION_TOKENS,
+ SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS,
result.usage.output_tokens
);
span.setAttribute(
diff --git a/sdk/typescript/src/instrumentation/cohere/wrapper.ts b/sdk/typescript/src/instrumentation/cohere/wrapper.ts
index 7842a3a7..344bab5f 100644
--- a/sdk/typescript/src/instrumentation/cohere/wrapper.ts
+++ b/sdk/typescript/src/instrumentation/cohere/wrapper.ts
@@ -30,8 +30,8 @@ export default class CohereWrapper extends BaseWrapper {
);
span.setAttribute(
- SemanticConvention.GEN_AI_TYPE,
- SemanticConvention.GEN_AI_TYPE_EMBEDDING
+ SemanticConvention.GEN_AI_OPERATION,
+ SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING
);
const { dimensions, encoding_format = 'float', input, user, texts = [] } = args[0];
@@ -46,7 +46,7 @@ export default class CohereWrapper extends BaseWrapper {
// Request Params attributes : Start
- span.setAttribute(SemanticConvention.GEN_AI_REQUEST_EMBEDDING_FORMAT, encoding_format);
+ span.setAttribute(SemanticConvention.GEN_AI_REQUEST_ENCODING_FORMATS, encoding_format);
span.setAttribute(SemanticConvention.GEN_AI_REQUEST_EMBEDDING_DIMENSION, dimensions);
if (traceContent) {
span.setAttribute(SemanticConvention.GEN_AI_CONTENT_PROMPT, JSON.stringify(texts));
@@ -56,7 +56,7 @@ export default class CohereWrapper extends BaseWrapper {
span.setAttribute(SemanticConvention.GEN_AI_RESPONSE_ID, response.id);
span.setAttribute(
- SemanticConvention.GEN_AI_USAGE_PROMPT_TOKENS,
+ SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
response.meta.billedUnits.inputTokens
);
span.setAttribute(
@@ -236,7 +236,7 @@ export default class CohereWrapper extends BaseWrapper {
}
// Request Params attributes : End
- span.setAttribute(SemanticConvention.GEN_AI_TYPE, SemanticConvention.GEN_AI_TYPE_CHAT);
+ span.setAttribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT);
span.setAttribute(SemanticConvention.GEN_AI_RESPONSE_ID, result.response_id);
@@ -259,11 +259,11 @@ export default class CohereWrapper extends BaseWrapper {
});
span.setAttribute(
- SemanticConvention.GEN_AI_USAGE_PROMPT_TOKENS,
+ SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
result.meta.billedUnits.inputTokens
);
span.setAttribute(
- SemanticConvention.GEN_AI_USAGE_COMPLETION_TOKENS,
+ SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS,
result.meta.billedUnits.outputTokens
);
span.setAttribute(
diff --git a/sdk/typescript/src/instrumentation/ollama/wrapper.ts b/sdk/typescript/src/instrumentation/ollama/wrapper.ts
index 879a4d61..3220113c 100644
--- a/sdk/typescript/src/instrumentation/ollama/wrapper.ts
+++ b/sdk/typescript/src/instrumentation/ollama/wrapper.ts
@@ -215,7 +215,7 @@ export default class OllamaWrapper {
}
const prompt = formattedMessages.join('\n');
- span.setAttribute(SemanticConvention.GEN_AI_TYPE, SemanticConvention.GEN_AI_TYPE_CHAT);
+ span.setAttribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT);
span.setAttribute(SemanticConvention.GEN_AI_RESPONSE_ID, result.id);
const model = result.model || args[0].model;
@@ -252,8 +252,8 @@ export default class OllamaWrapper {
}
// Request Params attributes : End
- span.setAttribute(SemanticConvention.GEN_AI_USAGE_PROMPT_TOKENS, promptTokens);
- span.setAttribute(SemanticConvention.GEN_AI_USAGE_COMPLETION_TOKENS, completionTokens);
+ span.setAttribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, promptTokens);
+ span.setAttribute(SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS, completionTokens);
span.setAttribute(SemanticConvention.GEN_AI_USAGE_TOTAL_TOKENS, totalTokens);
if (result.done_reason) {
diff --git a/sdk/typescript/src/instrumentation/openai/wrapper.ts b/sdk/typescript/src/instrumentation/openai/wrapper.ts
index fa6d5b77..54a458cf 100644
--- a/sdk/typescript/src/instrumentation/openai/wrapper.ts
+++ b/sdk/typescript/src/instrumentation/openai/wrapper.ts
@@ -219,7 +219,7 @@ export default class OpenAIWrapper extends BaseWrapper {
}
// Request Params attributes : End
- span.setAttribute(SemanticConvention.GEN_AI_TYPE, SemanticConvention.GEN_AI_TYPE_CHAT);
+ span.setAttribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_CHAT);
span.setAttribute(SemanticConvention.GEN_AI_RESPONSE_ID, result.id);
@@ -243,9 +243,9 @@ export default class OpenAIWrapper extends BaseWrapper {
aiSystem: OpenAIWrapper.aiSystem,
});
- span.setAttribute(SemanticConvention.GEN_AI_USAGE_PROMPT_TOKENS, result.usage.prompt_tokens);
+ span.setAttribute(SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS, result.usage.prompt_tokens);
span.setAttribute(
- SemanticConvention.GEN_AI_USAGE_COMPLETION_TOKENS,
+ SemanticConvention.GEN_AI_USAGE_OUTPUT_TOKENS,
result.usage.completion_tokens
);
span.setAttribute(SemanticConvention.GEN_AI_USAGE_TOTAL_TOKENS, result.usage.total_tokens);
@@ -298,8 +298,8 @@ export default class OpenAIWrapper extends BaseWrapper {
);
span.setAttribute(
- SemanticConvention.GEN_AI_TYPE,
- SemanticConvention.GEN_AI_TYPE_EMBEDDING
+ SemanticConvention.GEN_AI_OPERATION,
+ SemanticConvention.GEN_AI_OPERATION_TYPE_EMBEDDING
);
const { dimensions, encoding_format = 'float', input, user } = args[0];
@@ -314,7 +314,7 @@ export default class OpenAIWrapper extends BaseWrapper {
// Request Params attributes : Start
- span.setAttribute(SemanticConvention.GEN_AI_REQUEST_EMBEDDING_FORMAT, encoding_format);
+ span.setAttribute(SemanticConvention.GEN_AI_REQUEST_ENCODING_FORMATS, encoding_format);
span.setAttribute(SemanticConvention.GEN_AI_REQUEST_EMBEDDING_DIMENSION, dimensions);
if (traceContent) {
span.setAttribute(SemanticConvention.GEN_AI_CONTENT_PROMPT, input);
@@ -322,7 +322,7 @@ export default class OpenAIWrapper extends BaseWrapper {
// Request Params attributes : End
span.setAttribute(
- SemanticConvention.GEN_AI_USAGE_PROMPT_TOKENS,
+ SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens
);
span.setAttribute(
@@ -369,8 +369,8 @@ export default class OpenAIWrapper extends BaseWrapper {
});
span.setAttribute(
- SemanticConvention.GEN_AI_TYPE,
- SemanticConvention.GEN_AI_TYPE_FINETUNING
+ SemanticConvention.GEN_AI_OPERATION,
+ SemanticConvention.GEN_AI_OPERATION_TYPE_FINETUNING
);
// Request Params attributes : Start
@@ -394,7 +394,7 @@ export default class OpenAIWrapper extends BaseWrapper {
span.setAttribute(SemanticConvention.GEN_AI_RESPONSE_ID, response.id);
span.setAttribute(
- SemanticConvention.GEN_AI_USAGE_PROMPT_TOKENS,
+ SemanticConvention.GEN_AI_USAGE_INPUT_TOKENS,
response.usage.prompt_tokens
);
span.setAttribute(SemanticConvention.GEN_AI_REQUEST_FINETUNE_STATUS, response.status);
@@ -429,7 +429,7 @@ export default class OpenAIWrapper extends BaseWrapper {
user,
} = args[0];
- span.setAttribute(SemanticConvention.GEN_AI_TYPE, SemanticConvention.GEN_AI_TYPE_IMAGE);
+ span.setAttribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_IMAGE);
const model = response.model || 'dall-e-2';
@@ -504,7 +504,7 @@ export default class OpenAIWrapper extends BaseWrapper {
user,
} = args[0];
- span.setAttribute(SemanticConvention.GEN_AI_TYPE, SemanticConvention.GEN_AI_TYPE_IMAGE);
+ span.setAttribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_IMAGE);
span.setAttribute(SemanticConvention.GEN_AI_RESPONSE_ID, response.created);
const model = response.model || 'dall-e-2';
@@ -572,7 +572,7 @@ export default class OpenAIWrapper extends BaseWrapper {
const { input, user, voice, response_format = 'mp3', speed = 1 } = args[0];
- span.setAttribute(SemanticConvention.GEN_AI_TYPE, SemanticConvention.GEN_AI_TYPE_AUDIO);
+ span.setAttribute(SemanticConvention.GEN_AI_OPERATION, SemanticConvention.GEN_AI_OPERATION_TYPE_AUDIO);
const model = response.model || 'tts-1';
diff --git a/sdk/typescript/src/semantic-convention.ts b/sdk/typescript/src/semantic-convention.ts
index 34e59357..5ce460f6 100644
--- a/sdk/typescript/src/semantic-convention.ts
+++ b/sdk/typescript/src/semantic-convention.ts
@@ -4,7 +4,7 @@ export default class SemanticConvention {
static GEN_AI_SYSTEM = 'gen_ai.system';
static GEN_AI_ENVIRONMENT = 'gen_ai.environment';
static GEN_AI_APPLICATION_NAME = 'gen_ai.application_name';
- static GEN_AI_TYPE = 'gen_ai.type';
+ static GEN_AI_OPERATION = 'gen_ai.type';
static GEN_AI_HUB_OWNER = 'gen_ai.hub.owner';
static GEN_AI_HUB_REPO = 'gen_ai.hub.repo';
static GEN_AI_RETRIEVAL_SOURCE = 'gen_ai.retrieval.source';
@@ -21,7 +21,7 @@ export default class SemanticConvention {
static GEN_AI_REQUEST_SEED = 'gen_ai.request.seed';
static GEN_AI_REQUEST_FREQUENCY_PENALTY = 'gen_ai.request.frequency_penalty';
static GEN_AI_REQUEST_PRESENCE_PENALTY = 'gen_ai.request.presence_penalty';
- static GEN_AI_REQUEST_EMBEDDING_FORMAT = 'gen_ai.request.embedding_format';
+ static GEN_AI_REQUEST_ENCODING_FORMATS = 'gen_ai.request.embedding_format';
static GEN_AI_REQUEST_EMBEDDING_DIMENSION = 'gen_ai.request.embedding_dimension';
static GEN_AI_REQUEST_TOOL_CHOICE = 'gen_ai.request.tool_choice';
static GEN_AI_REQUEST_AUDIO_VOICE = 'gen_ai.request.audio_voice';
@@ -40,8 +40,8 @@ export default class SemanticConvention {
static GEN_AI_REQUEST_IMAGE_STYLE = 'gen_ai.request.image_style';
// GenAI Usage
- static GEN_AI_USAGE_PROMPT_TOKENS = 'gen_ai.usage.input_tokens';
- static GEN_AI_USAGE_COMPLETION_TOKENS = 'gen_ai.usage.output_tokens';
+ static GEN_AI_USAGE_INPUT_TOKENS = 'gen_ai.usage.input_tokens';
+ static GEN_AI_USAGE_OUTPUT_TOKENS = 'gen_ai.usage.output_tokens';
static GEN_AI_USAGE_TOTAL_TOKENS = 'gen_ai.usage.total_tokens';
static GEN_AI_USAGE_COST = 'gen_ai.usage.cost';
@@ -58,13 +58,13 @@ export default class SemanticConvention {
static GEN_AI_CONTENT_COMPLETION = 'gen_ai.completion';
static GEN_AI_CONTENT_REVISED_PROMPT = 'gen_ai.content.revised_prompt';
- static GEN_AI_TYPE_CHAT = 'chat';
- static GEN_AI_TYPE_EMBEDDING = 'embedding';
- static GEN_AI_TYPE_IMAGE = 'image';
- static GEN_AI_TYPE_AUDIO = 'audio';
- static GEN_AI_TYPE_FINETUNING = 'fine_tuning';
- static GEN_AI_TYPE_VECTORDB = 'vectordb';
- static GEN_AI_TYPE_FRAMEWORK = 'framework';
+ static GEN_AI_OPERATION_TYPE_CHAT = 'chat';
+ static GEN_AI_OPERATION_TYPE_EMBEDDING = 'embedding';
+ static GEN_AI_OPERATION_TYPE_IMAGE = 'image';
+ static GEN_AI_OPERATION_TYPE_AUDIO = 'audio';
+ static GEN_AI_OPERATION_TYPE_FINETUNING = 'fine_tuning';
+ static GEN_AI_OPERATION_TYPE_VECTORDB = 'vectordb';
+ static GEN_AI_OPERATION_TYPE_FRAMEWORK = 'framework';
static GEN_AI_SYSTEM_HUGGING_FACE = 'huggingface';
static GEN_AI_SYSTEM_OPENAI = 'openai';
@@ -72,7 +72,7 @@ export default class SemanticConvention {
static GEN_AI_SYSTEM_ANTHROPIC = 'anthropic';
static GEN_AI_SYSTEM_COHERE = 'cohere';
static GEN_AI_SYSTEM_MISTRAL = 'mistral';
- static GEN_AI_SYSTEM_BEDROCK = 'bedrock';
+ static GEN_AI_SYSTEM_AWS_BEDROCK = 'bedrock';
static GEN_AI_SYSTEM_VERTEXAI = 'vertexai';
static GEN_AI_SYSTEM_LANGCHAIN = 'langchain';