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]: Capture LLM Prompts and Responses in Trace Events (#354)

patcher9ae200425c58d
diff --git a/sdk/python/pyproject.toml b/sdk/python/pyproject.toml
index 73ebfb7c..f01a8b6e 100644
--- a/sdk/python/pyproject.toml
+++ b/sdk/python/pyproject.toml
@@ -1,6 +1,6 @@
[tool.poetry]
name = "openlit"
-version = "1.16.2"
+version = "1.17.0"
description = "OpenTelemetry-native Auto instrumentation library for monitoring LLM Applications, facilitating the integration of observability into your GenAI-driven projects"
authors = ["OpenLIT"]
repository = "https://github.com/openlit/openlit/tree/main/openlit/python"
diff --git a/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py b/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py
index a0a2778f..4cda8573 100644
--- a/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py
+++ b/sdk/python/src/openlit/instrumentation/anthropic/anthropic.py
@@ -130,7 +130,7 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_K,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- finish_reason)
+ [finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -140,10 +140,18 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -234,7 +242,7 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_K,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response.stop_reason)
+ [response.stop_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
response.usage.input_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -244,11 +252,21 @@ def messages(gen_ai_endpoint, version, environment, application_name, tracer,
response.usage.output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
+
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.content[0].text if response.content else "")
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.content[0].text if response.content else "",
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py b/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py
index a3227342..4f822141 100644
--- a/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py
+++ b/sdk/python/src/openlit/instrumentation/anthropic/async_anthropic.py
@@ -130,7 +130,7 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_K,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- finish_reason)
+ [finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -140,10 +140,18 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -234,7 +242,7 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_TOP_K,
kwargs.get("top_k", ""))
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response.stop_reason)
+ [response.stop_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
response.usage.input_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -245,10 +253,19 @@ def async_messages(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.content[0].text if response.content else "")
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.content[0].text if response.content else "",
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py b/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py
index bca07c78..0d710445 100644
--- a/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py
+++ b/sdk/python/src/openlit/instrumentation/bedrock/bedrock.py
@@ -96,7 +96,6 @@ def converse(gen_ai_endpoint, version, environment, application_name, tracer,
try:
message_prompt = method_kwargs.get("messages", "")
- print(message_prompt)
formatted_messages = []
for message in message_prompt:
role = message["role"]
@@ -144,10 +143,19 @@ def converse(gen_ai_endpoint, version, environment, application_name, tracer,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response["output"]["message"]["content"][0]["text"])
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response["output"]["message"]["content"][0]["text"],
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/cohere/cohere.py b/sdk/python/src/openlit/instrumentation/cohere/cohere.py
index e3bcd78f..b96b3c07 100644
--- a/sdk/python/src/openlit/instrumentation/cohere/cohere.py
+++ b/sdk/python/src/openlit/instrumentation/cohere/cohere.py
@@ -89,8 +89,12 @@ def embed(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -205,7 +209,7 @@ def chat(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response.generation_id)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response.finish_reason)
+ [response.finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
response.meta.billed_units.input_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -215,11 +219,20 @@ def chat(gen_ai_endpoint, version, environment, application_name, tracer,
response.meta.billed_units.output_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
+
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("message", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.text)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("message", ""),
+ },
+ )
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -348,7 +361,7 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_ID,
response_id)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- finish_reason)
+ [finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -358,10 +371,18 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("message", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("message", ""),
+ },
+ )
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: llmresponse,
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py b/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py
index 11ed8859..36a8b586 100644
--- a/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py
+++ b/sdk/python/src/openlit/instrumentation/elevenlabs/async_elevenlabs.py
@@ -80,8 +80,12 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- str(kwargs.get("text", "")))
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: str(kwargs.get("text", "")),
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py b/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py
index 235c0463..536a0b8b 100644
--- a/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py
+++ b/sdk/python/src/openlit/instrumentation/elevenlabs/elevenlabs.py
@@ -86,8 +86,12 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- str(kwargs.get("text", "")))
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: str(kwargs.get("text", "")),
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py b/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py
index 1dc764b8..9c6b5494 100644
--- a/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py
+++ b/sdk/python/src/openlit/instrumentation/gpt4all/gpt4all.py
@@ -107,10 +107,18 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
prompt_tokens)
@@ -195,10 +203,18 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response)
+ 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: response,
+ },
+ )
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
prompt_tokens)
@@ -313,8 +329,12 @@ def embed(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/groq/async_groq.py b/sdk/python/src/openlit/instrumentation/groq/async_groq.py
index d269bd8d..48434c88 100644
--- a/sdk/python/src/openlit/instrumentation/groq/async_groq.py
+++ b/sdk/python/src/openlit/instrumentation/groq/async_groq.py
@@ -141,10 +141,18 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -237,8 +245,12 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ 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:
@@ -254,23 +266,31 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].message.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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].message.content)
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[i].message.content,
+ },
+ )
i += 1
# Return original response
diff --git a/sdk/python/src/openlit/instrumentation/groq/groq.py b/sdk/python/src/openlit/instrumentation/groq/groq.py
index 9ccc7ecc..d057d937 100644
--- a/sdk/python/src/openlit/instrumentation/groq/groq.py
+++ b/sdk/python/src/openlit/instrumentation/groq/groq.py
@@ -141,10 +141,18 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -237,8 +245,12 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ 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:
@@ -254,23 +266,31 @@ def chat(gen_ai_endpoint, version, environment, application_name,
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].message.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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].message.content)
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[i].message.content,
+ },
+ )
i += 1
# Return original response
diff --git a/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py b/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py
index 28621165..9603711e 100644
--- a/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py
+++ b/sdk/python/src/openlit/instrumentation/mistral/async_mistral.py
@@ -103,7 +103,7 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response.choices[0].finish_reason)
+ [response.choices[0].finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
response.usage.prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -113,10 +113,19 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.choices[0].message.content if response.choices[0].message.content else "")
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[0].message.content if response.choices[0].message.content else "",
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -260,7 +269,7 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- finish_reason)
+ [finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -270,10 +279,18 @@ def async_chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -378,8 +395,12 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/mistral/mistral.py b/sdk/python/src/openlit/instrumentation/mistral/mistral.py
index 8a7453c5..983d63eb 100644
--- a/sdk/python/src/openlit/instrumentation/mistral/mistral.py
+++ b/sdk/python/src/openlit/instrumentation/mistral/mistral.py
@@ -102,7 +102,7 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response.choices[0].finish_reason)
+ [response.choices[0].finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
response.usage.prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -112,10 +112,19 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.choices[0].message.content if response.choices[0].message.content else "")
+ 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,
+ # pylint: disable=line-too-long
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.choices[0].message.content if response.choices[0].message.content else "",
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -259,7 +268,7 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
True)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- finish_reason)
+ [finish_reason])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
prompt_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
@@ -269,10 +278,18 @@ def chat_stream(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -377,8 +394,12 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py b/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py
index dfd74fdf..f97294a3 100644
--- a/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py
+++ b/sdk/python/src/openlit/instrumentation/ollama/async_ollama.py
@@ -120,10 +120,18 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -198,10 +206,18 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response['message']['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: response['message']['content'],
+ },
+ )
# Calculate cost of the operation
cost = 0
@@ -216,7 +232,7 @@ def async_chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response["done_reason"])
+ [response["done_reason"]])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
@@ -345,10 +361,19 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
+ },
+ )
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: llmresponse,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -404,10 +429,18 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response['response'])
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
+ },
+ )
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response['response'],
+ },
+ )
# Calculate cost of the operation
cost = 0
@@ -422,7 +455,7 @@ def async_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response["done_reason"])
+ [response["done_reason"]])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
@@ -525,8 +558,12 @@ def async_embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get('prompt', ""))
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/ollama/ollama.py b/sdk/python/src/openlit/instrumentation/ollama/ollama.py
index 78e94e30..79071d3f 100644
--- a/sdk/python/src/openlit/instrumentation/ollama/ollama.py
+++ b/sdk/python/src/openlit/instrumentation/ollama/ollama.py
@@ -120,10 +120,18 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -198,10 +206,18 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response['message']['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: response['message']['content'],
+ },
+ )
# Calculate cost of the operation
cost = 0
@@ -216,7 +232,7 @@ def chat(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response["done_reason"])
+ [response["done_reason"]])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
@@ -345,10 +361,19 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
+ },
+ )
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: llmresponse,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -404,10 +429,18 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response['response'])
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
+ },
+ )
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response['response'],
+ },
+ )
# Calculate cost of the operation
cost = 0
@@ -422,7 +455,7 @@ def generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_TOTAL_TOKENS,
total_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_RESPONSE_FINISH_REASON,
- response["done_reason"])
+ [response["done_reason"]])
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
@@ -525,8 +558,12 @@ def embeddings(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get('prompt', ""))
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: kwargs.get("prompt", ""),
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/openai/async_azure_openai.py b/sdk/python/src/openlit/instrumentation/openai/async_azure_openai.py
index f909b3e2..16704f98 100644
--- a/sdk/python/src/openlit/instrumentation/openai/async_azure_openai.py
+++ b/sdk/python/src/openlit/instrumentation/openai/async_azure_openai.py
@@ -143,10 +143,18 @@ def azure_async_chat_completions(gen_ai_endpoint, version, environment, applicat
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -238,8 +246,12 @@ def azure_async_chat_completions(gen_ai_endpoint, version, environment, applicat
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ 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:
@@ -255,23 +267,31 @@ def azure_async_chat_completions(gen_ai_endpoint, version, environment, applicat
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].message.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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].message.content)
+ 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
@@ -284,8 +304,12 @@ def azure_async_chat_completions(gen_ai_endpoint, version, environment, applicat
response.usage.prompt_tokens,
response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- "Function called with tools")
+ 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,
@@ -443,10 +467,18 @@ def azure_async_completions(gen_ai_endpoint, version, environment, application_n
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -519,8 +551,12 @@ def azure_async_completions(gen_ai_endpoint, version, environment, application_n
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
+ 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:
@@ -536,23 +572,31 @@ def azure_async_completions(gen_ai_endpoint, version, environment, application_n
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].text)
+ 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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].text)
+ 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
@@ -563,8 +607,12 @@ def azure_async_completions(gen_ai_endpoint, version, environment, application_n
response.usage.prompt_tokens,
response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- "Function called with tools")
+ 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,
@@ -679,8 +727,12 @@ def azure_async_embedding(gen_ai_endpoint, version, environment, application_nam
response.usage.total_tokens)
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST, cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("input", ""))
+ 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))
@@ -796,12 +848,19 @@ def azure_async_image_generate(gen_ai_endpoint, version, environment, applicatio
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
-
+ 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.set_attribute(attribute_name,
- getattr(items, image))
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
+ },
+ )
images_count+=1
diff --git a/sdk/python/src/openlit/instrumentation/openai/async_openai.py b/sdk/python/src/openlit/instrumentation/openai/async_openai.py
index 0d7d43ec..2768ea22 100644
--- a/sdk/python/src/openlit/instrumentation/openai/async_openai.py
+++ b/sdk/python/src/openlit/instrumentation/openai/async_openai.py
@@ -144,10 +144,18 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -240,8 +248,12 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -259,23 +271,31 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].message.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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].message.content)
+ 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
@@ -288,8 +308,12 @@ def async_chat_completions(gen_ai_endpoint, version, environment, application_na
pricing_info, response.usage.prompt_tokens,
response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- "Function called with tools")
+ 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,
@@ -405,8 +429,12 @@ def async_embedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("input", ""))
+ 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))
@@ -611,12 +639,19 @@ def async_image_generate(gen_ai_endpoint, version, environment, application_name
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
-
+ 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.set_attribute(attribute_name,
- getattr(items, image))
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
+ },
+ )
images_count+=1
@@ -729,11 +764,13 @@ def async_image_variatons(gen_ai_endpoint, version, environment, application_nam
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
"standard")
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get(SemanticConvetion.GEN_AI_TYPE_IMAGE, ""))
-
attribute_name = f"gen_ai.response.image.{images_count}"
- span.set_attribute(attribute_name, getattr(items, image))
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
+ },
+ )
images_count+=1
@@ -838,8 +875,12 @@ def async_audio_create(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("input", ""))
+ 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))
diff --git a/sdk/python/src/openlit/instrumentation/openai/azure_openai.py b/sdk/python/src/openlit/instrumentation/openai/azure_openai.py
index fd02aea2..eebd9b69 100644
--- a/sdk/python/src/openlit/instrumentation/openai/azure_openai.py
+++ b/sdk/python/src/openlit/instrumentation/openai/azure_openai.py
@@ -143,10 +143,18 @@ def azure_chat_completions(gen_ai_endpoint, version, environment, application_na
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -237,6 +245,13 @@ def azure_chat_completions(gen_ai_endpoint, version, environment, application_na
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)
@@ -255,23 +270,31 @@ def azure_chat_completions(gen_ai_endpoint, version, environment, application_na
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].message.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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].message.content)
+ 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
@@ -284,8 +307,12 @@ def azure_chat_completions(gen_ai_endpoint, version, environment, application_na
response.usage.prompt_tokens,
response.usage.completion_tokens)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- "Function called with tools")
+ 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,
@@ -441,10 +468,18 @@ def azure_completions(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -517,8 +552,12 @@ def azure_completions(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
+ 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:
@@ -534,23 +573,31 @@ def azure_completions(gen_ai_endpoint, version, environment, application_name,
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].text)
+ 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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].text)
+ 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
@@ -678,8 +725,12 @@ def azure_embedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("input", ""))
+ 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))
@@ -795,12 +846,19 @@ def azure_image_generate(gen_ai_endpoint, version, environment, application_name
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
-
+ 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.set_attribute(attribute_name,
- getattr(items, image))
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
+ },
+ )
images_count+=1
diff --git a/sdk/python/src/openlit/instrumentation/openai/openai.py b/sdk/python/src/openlit/instrumentation/openai/openai.py
index 93b01013..3ee03f6b 100644
--- a/sdk/python/src/openlit/instrumentation/openai/openai.py
+++ b/sdk/python/src/openlit/instrumentation/openai/openai.py
@@ -145,10 +145,18 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -241,8 +249,12 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IS_STREAM,
False)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ 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:
@@ -258,23 +270,31 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
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)
+ [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.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.choices[0].message.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.completion.{i}"
- span.set_attribute(attribute_name,
- response.choices[i].message.content)
+ 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
@@ -286,9 +306,12 @@ def chat_completions(gen_ai_endpoint, version, environment, application_name,
cost = get_chat_model_cost(kwargs.get("model", "gpt-3.5-turbo"),
pricing_info, response.usage.prompt_tokens,
response.usage.completion_tokens)
-
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- "Function called with tools")
+ 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,
@@ -404,8 +427,12 @@ def embedding(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("input", ""))
+ 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))
@@ -626,12 +653,19 @@ def image_generate(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_USER,
kwargs.get("user", ""))
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("prompt", ""))
-
+ 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.set_attribute(attribute_name,
- getattr(items, image))
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
+ },
+ )
images_count+=1
@@ -744,12 +778,13 @@ def image_variatons(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_IMAGE_QUALITY,
"standard")
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get(SemanticConvetion.GEN_AI_TYPE_IMAGE, ""))
-
attribute_name = f"gen_ai.response.image.{images_count}"
- span.set_attribute(attribute_name,
- getattr(items, image))
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: getattr(items, image),
+ },
+ )
images_count+=1
@@ -854,8 +889,12 @@ def audio_create(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- kwargs.get("input", ""))
+ 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))
diff --git a/sdk/python/src/openlit/instrumentation/transformers/transformers.py b/sdk/python/src/openlit/instrumentation/transformers/transformers.py
index e96e61f9..16a0c8f0 100644
--- a/sdk/python/src/openlit/instrumentation/transformers/transformers.py
+++ b/sdk/python/src/openlit/instrumentation/transformers/transformers.py
@@ -88,29 +88,33 @@ def text_wrap(gen_ai_endpoint, version, environment, application_name,
forward_params.get("top_p", "null"))
span.set_attribute(SemanticConvetion.GEN_AI_REQUEST_MAX_TOKENS,
forward_params.get("max_length", -1))
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
- prompt_tokens)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
+ span.set_attribute(SemanticConvetion.GEN_AI_USAGE_PROMPT_TOKENS,
+ prompt_tokens)
i = 0
completion_tokens = 0
for completion in response:
if len(response) > 1:
- attribute_name = f"gen_ai.completion.{i}"
+ attribute_name = f"gen_ai.content.completion.{i}"
else:
- attribute_name = SemanticConvetion.GEN_AI_CONTENT_COMPLETION
- if i == 0:
- if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- completion["generated_text"])
- completion_tokens += general_tokens(completion["generated_text"])
- else:
- if trace_content:
- span.set_attribute(attribute_name,
- completion["generated_text"])
- completion_tokens += general_tokens(completion["generated_text"])
+ attribute_name = SemanticConvetion.GEN_AI_CONTENT_COMPLETION_EVENT
+ if trace_content:
+ span.add_event(
+ name=attribute_name,
+ attributes={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: completion["generated_text"],
+ },
+ )
+ completion_tokens += general_tokens(completion["generated_text"])
+
i=i+1
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COMPLETION_TOKENS,
completion_tokens)
diff --git a/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py b/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py
index 91ec4bd7..b1d0f651 100644
--- a/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py
+++ b/sdk/python/src/openlit/instrumentation/vertexai/async_vertexai.py
@@ -103,10 +103,18 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -181,10 +189,19 @@ def generate_content_async(gen_ai_endpoint, version, environment, application_na
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.candidates[0].content.parts[0].text)
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.candidates[0].content.parts[0].text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -316,10 +333,18 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -394,11 +419,19 @@ def send_message_async(gen_ai_endpoint, version, environment, application_name,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.candidates[0].content.parts[0].text)
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.candidates[0].content.parts[0].text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -516,10 +549,18 @@ def predict_async(gen_ai_endpoint, version, environment, application_name, trace
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.text)
+ 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: response.text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -648,10 +689,18 @@ def predict_streaming_async(gen_ai_endpoint, version, environment, application_n
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -765,10 +814,18 @@ def start_chat_async(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.text)
+ 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: response.text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -895,10 +952,18 @@ def start_chat_streaming_async(gen_ai_endpoint, version, environment, applicatio
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -1006,8 +1071,12 @@ def embeddings_async(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py b/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py
index 2657dfae..c0da193f 100644
--- a/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py
+++ b/sdk/python/src/openlit/instrumentation/vertexai/vertexai.py
@@ -103,10 +103,18 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -181,10 +189,19 @@ def generate_content(gen_ai_endpoint, version, environment, application_name, tr
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.candidates[0].content.parts[0].text)
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.candidates[0].content.parts[0].text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -316,10 +333,18 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -394,11 +419,19 @@ def send_message(gen_ai_endpoint, version, environment, application_name, tracer
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- response.candidates[0].content.parts[0].text)
+ 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={
+ # pylint: disable=line-too-long
+ SemanticConvetion.GEN_AI_CONTENT_COMPLETION: response.candidates[0].content.parts[0].text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -516,10 +549,18 @@ def predict(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.text)
+ 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: response.text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -648,10 +689,18 @@ def predict_streaming(gen_ai_endpoint, version, environment, application_name, t
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -765,10 +814,18 @@ def start_chat(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- # pylint: disable=line-too-long
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION, response.text)
+ 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: response.text,
+ },
+ )
span.set_status(Status(StatusCode.OK))
@@ -895,10 +952,18 @@ def start_chat_streaming(gen_ai_endpoint, version, environment, application_name
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_COMPLETION,
- llmresponse)
+ 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))
@@ -1006,8 +1071,12 @@ def embeddings(gen_ai_endpoint, version, environment, application_name, tracer,
span.set_attribute(SemanticConvetion.GEN_AI_USAGE_COST,
cost)
if trace_content:
- span.set_attribute(SemanticConvetion.GEN_AI_CONTENT_PROMPT,
- prompt)
+ span.add_event(
+ name=SemanticConvetion.GEN_AI_CONTENT_PROMPT_EVENT,
+ attributes={
+ SemanticConvetion.GEN_AI_CONTENT_PROMPT: prompt,
+ },
+ )
span.set_status(Status(StatusCode.OK))
diff --git a/sdk/python/src/openlit/semcov/__init__.py b/sdk/python/src/openlit/semcov/__init__.py
index f527244c..34682538 100644
--- a/sdk/python/src/openlit/semcov/__init__.py
+++ b/sdk/python/src/openlit/semcov/__init__.py
@@ -70,7 +70,9 @@ class SemanticConvetion:
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"
@@ -94,7 +96,7 @@ class SemanticConvetion:
GEN_AI_SYSTEM_COHERE = "cohere"
GEN_AI_SYSTEM_MISTRAL = "mistral"
GEN_AI_SYSTEM_BEDROCK = "bedrock"
- GEN_AI_SYSTEM_VERTEXAI = "vertexai"
+ GEN_AI_SYSTEM_VERTEXAI = "vertex_ai"
GEN_AI_SYSTEM_GROQ = "groq"
GEN_AI_SYSTEM_OLLAMA = "ollama"
GEN_AI_SYSTEM_GPT4ALL = "gpt4all"
diff --git a/src/client/src/constants/traces.ts b/src/client/src/constants/traces.ts
index 7906c289..a3f6340a 100644
--- a/src/client/src/constants/traces.ts
+++ b/src/client/src/constants/traces.ts
@@ -264,8 +264,8 @@ export const TraceMapping: Record<
prompt: {
label: "Prompt",
type: "string",
- path: "prompt",
- prefix: SpanAttributesGenAIPrefix,
+ path: ["Events.Attributes", "0", "gen_ai.prompt"],
+ isRoot: true,
},
finishReason: {
label: "Finish Reason",
@@ -276,8 +276,8 @@ export const TraceMapping: Record<
response: {
label: "Response",
type: "string",
- path: "completion",
- prefix: SpanAttributesGenAIPrefix,
+ path: ["Events.Attributes", "1", "gen_ai.completion"],
+ isRoot: true,
},
randomSeed: {
label: "Random seed",
@@ -473,4 +473,4 @@ export interface TraceRow {
TraceState: string;
Attributes: Record<string, string>;
}[];
-}
\ No newline at end of file
+}