Skip to content

Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.

fix: use provider instead of backend

qin-ctx69e141fd88c5
diff --git a/README.md b/README.md
index 6b92629b1..dd8ab82d5 100644
--- a/README.md
+++ b/README.md
@@ -92,7 +92,7 @@ Create a configuration file `ov.conf`:
"dense": {
"api_base" : "<api-endpoint>", // API endpoint address
"api_key" : "<your-api-key>", // Model service API Key
- "backend" : "<backend-type>", // Backend type (volcengine or openai)
+ "provider" : "<provider-type>", // Provider type (volcengine or openai)
"dimension": 1024, // Vector dimension
"model" : "<model-name>" // Embedding model name (e.g., doubao-embedding-vision-250615 or text-embedding-3-large)
}
@@ -100,7 +100,7 @@ Create a configuration file `ov.conf`:
"vlm": {
"api_base" : "<api-endpoint>", // API endpoint address
"api_key" : "<your-api-key>", // Model service API Key
- "backend" : "<backend-type>", // Backend type (volcengine or openai)
+ "provider" : "<provider-type>", // Provider type (volcengine or openai)
"model" : "<model-name>" // VLM model name (e.g., doubao-seed-1-8-251228 or gpt-4-vision-preview)
}
}
@@ -119,7 +119,7 @@ Create a configuration file `ov.conf`:
"dense": {
"api_base" : "https://ark.cn-beijing.volces.com/api/v3",
"api_key" : "your-volcengine-api-key",
- "backend" : "volcengine",
+ "provider" : "volcengine",
"dimension": 1024,
"model" : "doubao-embedding-vision-250615"
}
@@ -127,7 +127,7 @@ Create a configuration file `ov.conf`:
"vlm": {
"api_base" : "https://ark.cn-beijing.volces.com/api/v3",
"api_key" : "your-volcengine-api-key",
- "backend" : "volcengine",
+ "provider" : "volcengine",
"model" : "doubao-seed-1-8-251228"
}
}
@@ -144,7 +144,7 @@ Create a configuration file `ov.conf`:
"dense": {
"api_base" : "https://api.openai.com/v1",
"api_key" : "your-openai-api-key",
- "backend" : "openai",
+ "provider" : "openai",
"dimension": 3072,
"model" : "text-embedding-3-large"
}
@@ -152,7 +152,7 @@ Create a configuration file `ov.conf`:
"vlm": {
"api_base" : "https://api.openai.com/v1",
"api_key" : "your-openai-api-key",
- "backend" : "openai",
+ "provider" : "openai",
"model" : "gpt-4-vision-preview"
}
}
diff --git a/README_CN.md b/README_CN.md
index ad550a391..2167ac68a 100644
--- a/README_CN.md
+++ b/README_CN.md
@@ -91,7 +91,7 @@ OpenViking 支持多种模型服务:
"dense": {
"api_base" : "<api-endpoint>", // API 服务端点地址
"api_key" : "<your-api-key>", // 模型服务的 API 密钥
- "backend" : "<backend-type>", // 后端类型(volcengine 或 openai)
+ "provider" : "<provider-type>", // 提供商类型(volcengine 或 openai)
"dimension": 1024, // 向量维度
"model" : "<model-name>" // Embedding 模型名称(如 doubao-embedding-vision-250615 或 text-embedding-3-large)
}
@@ -99,7 +99,7 @@ OpenViking 支持多种模型服务:
"vlm": {
"api_base" : "<api-endpoint>", // API 服务端点地址
"api_key" : "<your-api-key>", // 模型服务的 API 密钥
- "backend" : "<backend-type>", // 后端类型(volcengine 或 openai)
+ "provider" : "<provider-type>", // 提供商类型(volcengine 或 openai)
"model" : "<model-name>" // VLM 模型名称(如 doubao-seed-1-8-251228 或 gpt-4-vision-preview)
}
}
@@ -118,7 +118,7 @@ OpenViking 支持多种模型服务:
"dense": {
"api_base" : "https://ark.cn-beijing.volces.com/api/v3",
"api_key" : "your-volcengine-api-key",
- "backend" : "volcengine",
+ "provider" : "volcengine",
"dimension": 1024,
"model" : "doubao-embedding-vision-250615"
}
@@ -126,7 +126,7 @@ OpenViking 支持多种模型服务:
"vlm": {
"api_base" : "https://ark.cn-beijing.volces.com/api/v3",
"api_key" : "your-volcengine-api-key",
- "backend" : "volcengine",
+ "provider" : "volcengine",
"model" : "doubao-seed-1-8-251228"
}
}
@@ -143,7 +143,7 @@ OpenViking 支持多种模型服务:
"dense": {
"api_base" : "https://api.openai.com/v1",
"api_key" : "your-openai-api-key",
- "backend" : "openai",
+ "provider" : "openai",
"dimension": 3072,
"model" : "text-embedding-3-large"
}
@@ -151,7 +151,7 @@ OpenViking 支持多种模型服务:
"vlm": {
"api_base" : "https://api.openai.com/v1",
"api_key" : "your-openai-api-key",
- "backend" : "openai",
+ "provider" : "openai",
"model" : "gpt-4-vision-preview"
}
}
diff --git a/docs/en/configuration/configuration.md b/docs/en/configuration/configuration.md
index 2150dbfd5..45ba3942b 100644
--- a/docs/en/configuration/configuration.md
+++ b/docs/en/configuration/configuration.md
@@ -10,13 +10,14 @@ Create `ov.conf` in your project directory:
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-embedding-vision-250615",
"dimension": 1024
}
},
"vlm": {
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-seed-1-8-251228"
},
@@ -48,7 +49,7 @@ Embedding model configuration for vector search.
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-embedding-vision-250615",
"dimension": 1024,
@@ -67,6 +68,7 @@ Vision Language Model for semantic extraction.
```json
{
"vlm": {
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-seed-1-8-251228",
"base_url": "https://ark.cn-beijing.volces.com/api/v3"
@@ -152,7 +154,7 @@ config = OpenVikingConfig(
),
embedding=EmbeddingConfig(
dense=DenseEmbeddingConfig(
- backend="volcengine",
+ provider="volcengine",
api_key="your-api-key",
model="doubao-embedding-vision-250615",
dimension=1024
@@ -171,7 +173,7 @@ client = ov.AsyncOpenViking(config=config)
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "string",
"model": "string",
"dimension": 1024,
@@ -179,6 +181,7 @@ client = ov.AsyncOpenViking(config=config)
}
},
"vlm": {
+ "provider": "string",
"api_key": "string",
"model": "string",
"base_url": "string"
diff --git a/docs/en/configuration/embedding.md b/docs/en/configuration/embedding.md
index b03e20f40..da58d9865 100644
--- a/docs/en/configuration/embedding.md
+++ b/docs/en/configuration/embedding.md
@@ -8,7 +8,7 @@ Configure embedding models for vector search.
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-volcengine-api-key",
"model": "doubao-embedding-vision-250615",
"dimension": 1024,
@@ -22,7 +22,7 @@ Configure embedding models for vector search.
| Parameter | Type | Description |
|-----------|------|-------------|
-| `backend` | str | `"volcengine"` |
+| `provider` | str | `"volcengine"` |
| `api_key` | str | Volcengine API key |
| `model` | str | Model name |
| `dimension` | int | Vector dimension |
@@ -54,7 +54,7 @@ Then in config:
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"model": "doubao-embedding-vision-250615",
"dimension": 1024
}
@@ -69,7 +69,7 @@ from openviking.utils.config import EmbeddingConfig, DenseEmbeddingConfig
embedding_config = EmbeddingConfig(
dense=DenseEmbeddingConfig(
- backend="volcengine",
+ provider="volcengine",
api_key="your-api-key",
model="doubao-embedding-vision-250615",
dimension=1024,
diff --git a/docs/en/configuration/llm.md b/docs/en/configuration/llm.md
index df7c10aa3..3fbe0fcb8 100644
--- a/docs/en/configuration/llm.md
+++ b/docs/en/configuration/llm.md
@@ -9,6 +9,7 @@ Used for generating L0/L1 content from resources.
```json
{
"vlm": {
+ "provider": "volcengine",
"api_key": "your-volcengine-api-key",
"model": "doubao-seed-1-8-251228",
"base_url": "https://ark.cn-beijing.volces.com/api/v3"
diff --git a/docs/zh/configuration/configuration.md b/docs/zh/configuration/configuration.md
index cc26e5eb9..87b1aec80 100644
--- a/docs/zh/configuration/configuration.md
+++ b/docs/zh/configuration/configuration.md
@@ -11,7 +11,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
"user": "default_user",
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-embedding-vision-250615",
"dimension": 1024,
@@ -19,7 +19,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
}
},
"vlm": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-seed-1-8-251228",
"api_base": "https://ark.cn-beijing.volces.com/api/v3"
@@ -59,7 +59,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-embedding-vision-250615",
"dimension": 1024,
@@ -70,18 +70,18 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
}
```
-**支持的 backend:**
+**支持的 provider:**
- `openai`: OpenAI Embedding API
- `volcengine`: 火山引擎 Embedding API
- `vikingdb`: VikingDB Embedding API
-**vikingdb backend 配置示例:**
+**vikingdb provider 配置示例:**
```json
{
"embedding": {
"dense": {
- "backend": "vikingdb",
+ "provider": "vikingdb",
"model": "bge_large_zh",
"ak": "your-access-key",
"sk": "your-secret-key",
@@ -98,7 +98,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
{
"embedding": {
"sparse": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "bm25-sparse-v1"
}
@@ -116,7 +116,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
{
"embedding": {
"hybrid": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-embedding-hybrid",
"dimension": 1024
@@ -131,13 +131,13 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-embedding-vision-250615",
"dimension": 1024
},
"sparse": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "bm25-sparse-v1"
}
@@ -154,6 +154,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支
```json
{
"vlm": {
+ "provider": "volcengine",
"api_key": "your-api-key",
"model": "doubao-seed-1-8-251228",
"base_url": "https://ark.cn-beijing.volces.com/api/v3"
@@ -239,7 +240,7 @@ config = OpenVikingConfig(
),
embedding=EmbeddingConfig(
dense=DenseEmbeddingConfig(
- backend="volcengine",
+ provider="volcengine",
api_key="your-api-key",
model="doubao-embedding-vision-250615",
dimension=1024
@@ -258,7 +259,7 @@ client = ov.AsyncOpenViking(config=config)
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "string",
"model": "string",
"dimension": 1024,
@@ -266,6 +267,7 @@ client = ov.AsyncOpenViking(config=config)
}
},
"vlm": {
+ "provider": "string",
"api_key": "string",
"model": "string",
"base_url": "string"
diff --git a/docs/zh/configuration/embedding.md b/docs/zh/configuration/embedding.md
index 49dd881c8..626ee5cbe 100644
--- a/docs/zh/configuration/embedding.md
+++ b/docs/zh/configuration/embedding.md
@@ -8,7 +8,7 @@
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"api_key": "your-volcengine-api-key",
"model": "doubao-embedding-vision-250615",
"dimension": 1024,
@@ -22,7 +22,7 @@
| 参数 | 类型 | 说明 |
|------|------|------|
-| `backend` | str | `"volcengine"` |
+| `provider` | str | `"volcengine"` |
| `api_key` | str | 火山引擎 API Key |
| `model` | str | 模型名称 |
| `dimension` | int | 向量维度 |
@@ -54,7 +54,7 @@ export VOLCENGINE_API_KEY="your-api-key"
{
"embedding": {
"dense": {
- "backend": "volcengine",
+ "provider": "volcengine",
"model": "doubao-embedding-vision-250615",
"dimension": 1024
}
@@ -69,7 +69,7 @@ from openviking.utils.config import EmbeddingConfig, DenseEmbeddingConfig
embedding_config = EmbeddingConfig(
dense=DenseEmbeddingConfig(
- backend="volcengine",
+ provider="volcengine",
api_key="your-api-key",
model="doubao-embedding-vision-250615",
dimension=1024,
diff --git a/docs/zh/configuration/llm.md b/docs/zh/configuration/llm.md
index ca8047d93..a8a714ddf 100644
--- a/docs/zh/configuration/llm.md
+++ b/docs/zh/configuration/llm.md
@@ -9,6 +9,7 @@
```json
{
"vlm": {
+ "provider": "volcengine",
"api_key": "your-volcengine-api-key",
"model": "doubao-seed-1-8-251228",
"base_url": "https://ark.cn-beijing.volces.com/api/v3"
diff --git a/examples/ov.conf.example b/examples/ov.conf.example
index b853b6ad9..353d94e73 100644
--- a/examples/ov.conf.example
+++ b/examples/ov.conf.example
@@ -29,7 +29,7 @@
"api_key": "{your-api-key}",
"api_base": "https://ark.cn-beijing.volces.com/api/v3",
"dimension": 1024,
- "backend": "volcengine",
+ "provider": "volcengine",
"input": "multimodal"
}
},
@@ -39,7 +39,7 @@
"api_base": "https://ark.cn-beijing.volces.com/api/v3",
"temperature": 0.0,
"max_retries": 2,
- "backend": "volcengine"
+ "provider": "volcengine"
},
"rerank": {
"ak": null,
diff --git a/openviking/models/vlm/backends/openai_vlm.py b/openviking/models/vlm/backends/openai_vlm.py
index 50054219f..27b014498 100644
--- a/openviking/models/vlm/backends/openai_vlm.py
+++ b/openviking/models/vlm/backends/openai_vlm.py
@@ -17,8 +17,8 @@ class OpenAIVLM(VLMBase):
super().__init__(config)
self._sync_client = None
self._async_client = None
- # Ensure backend type is correct
- self.backend = "openai"
+ # Ensure provider type is correct
+ self.provider = "openai"
def get_client(self):
"""Get sync client"""
@@ -46,7 +46,7 @@ class OpenAIVLM(VLMBase):
completion_tokens = response.usage.completion_tokens
self.update_token_usage(
model_name=self.model or "gpt-4o-mini",
- backend="openai",
+ provider=self.provider,
prompt_tokens=prompt_tokens,
completion_tokens=completion_tokens,
)
diff --git a/openviking/models/vlm/backends/volcengine_vlm.py b/openviking/models/vlm/backends/volcengine_vlm.py
index 3ee951707..8f1d2e97c 100644
--- a/openviking/models/vlm/backends/volcengine_vlm.py
+++ b/openviking/models/vlm/backends/volcengine_vlm.py
@@ -16,8 +16,8 @@ class VolcEngineVLM(OpenAIVLM):
super().__init__(config)
self._sync_client = None
self._async_client = None
- # Ensure backend type is correct
- self.backend = "volcengine"
+ # Ensure provider type is correct
+ self.provider = "volcengine"
# VolcEngine-specific defaults
if not self.api_base:
diff --git a/openviking/models/vlm/base.py b/openviking/models/vlm/base.py
index fc1c797e7..7111fe162 100644
--- a/openviking/models/vlm/base.py
+++ b/openviking/models/vlm/base.py
@@ -14,7 +14,8 @@ class VLMBase(ABC):
def __init__(self, config: Dict[str, Any]):
self.config = config
- self.backend = config.get("backend", "openai")
+ self.provider = config.get("backend", "openai")
+
self.model = config.get("model")
self.api_key = config.get("api_key")
self.api_base = config.get("api_base")
@@ -58,19 +59,19 @@ class VLMBase(ABC):
# Token usage tracking methods
def update_token_usage(
- self, model_name: str, backend: str, prompt_tokens: int, completion_tokens: int
+ self, model_name: str, provider: str, prompt_tokens: int, completion_tokens: int
) -> None:
"""Update token usage
Args:
model_name: Model name
- backend: Backend name (openai, volcengine)
+ provider: Provider name (openai, volcengine)
prompt_tokens: Number of prompt tokens
completion_tokens: Number of completion tokens
"""
self._token_tracker.update(
model_name=model_name,
- backend=backend,
+ provider=provider,
prompt_tokens=prompt_tokens,
completion_tokens=completion_tokens,
)
@@ -110,26 +111,27 @@ class VLMFactory:
"""Create VLM instance
Args:
- config: VLM config, must contain 'backend' field
+ config: VLM config, must contain 'provider' field
Returns:
VLMBase: VLM instance
Raises:
- ValueError: If backend is not supported
+ ValueError: If provider is not supported
ImportError: If related dependencies are not installed
"""
- backend = config.get("backend", "openai")
- if backend == "openai":
+ provider = config.get("backend", "openai")
+
+ if provider == "openai":
from .backends.openai_vlm import OpenAIVLM
return OpenAIVLM(config)
- elif backend == "volcengine":
+ elif provider == "volcengine":
from .backends.volcengine_vlm import VolcEngineVLM
return VolcEngineVLM(config)
else:
- raise ValueError(f"Unsupported VLM backend: {backend}")
+ raise ValueError(f"Unsupported VLM provider: {provider}")
@staticmethod
def get_available_backends() -> List[str]:
- """Get list of available backends"""
+ """Get list of available providers"""
return ["openai", "volcengine"]
diff --git a/openviking/models/vlm/token_usage.py b/openviking/models/vlm/token_usage.py
index 6b29f635d..f1b511946 100644
--- a/openviking/models/vlm/token_usage.py
+++ b/openviking/models/vlm/token_usage.py
@@ -62,35 +62,35 @@ class ModelTokenUsage:
model_name: str
total_usage: TokenUsage = field(default_factory=TokenUsage)
- usage_by_backend: Dict[str, TokenUsage] = field(default_factory=dict)
+ usage_by_provider: Dict[str, TokenUsage] = field(default_factory=dict)
- def update(self, backend: str, prompt_tokens: int, completion_tokens: int) -> None:
- """Update token usage for specified backend
+ def update(self, provider: str, prompt_tokens: int, completion_tokens: int) -> None:
+ """Update token usage for specified provider
Args:
- backend: Backend name (openai, volcengine)
+ provider: Provider name (openai, volcengine)
prompt_tokens: Number of input tokens
completion_tokens: Number of output tokens
"""
# Update total usage
self.total_usage.update(prompt_tokens, completion_tokens)
- # Update backend usage
- if backend not in self.usage_by_backend:
- self.usage_by_backend[backend] = TokenUsage()
+ # Update provider usage
+ if provider not in self.usage_by_provider:
+ self.usage_by_provider[provider] = TokenUsage()
- self.usage_by_backend[backend].update(prompt_tokens, completion_tokens)
+ self.usage_by_provider[provider].update(prompt_tokens, completion_tokens)
- def get_backend_usage(self, backend: str) -> Optional[TokenUsage]:
- """Get token usage for specified backend
+ def get_provider_usage(self, provider: str) -> Optional[TokenUsage]:
+ """Get token usage for specified provider
Args:
- backend: Backend name
+ provider: Provider name
Returns:
- TokenUsage object, or None if backend doesn't exist
+ TokenUsage object, or None if provider doesn't exist
"""
- return self.usage_by_backend.get(backend)
+ return self.usage_by_provider.get(provider)
def to_dict(self) -> Dict:
"""Convert to dictionary format
@@ -101,19 +101,19 @@ class ModelTokenUsage:
result = {
"model_name": self.model_name,
"total_usage": self.total_usage.to_dict(),
- "usage_by_backend": {},
+ "usage_by_provider": {},
}
- for backend, usage in self.usage_by_backend.items():
- result["usage_by_backend"][backend] = usage.to_dict()
+ for provider, usage in self.usage_by_provider.items():
+ result["usage_by_provider"][provider] = usage.to_dict()
return result
def __str__(self) -> str:
- backends = ", ".join(
- [f"{backend}: {usage.total_tokens}" for backend, usage in self.usage_by_backend.items()]
+ providers = ", ".join(
+ [f"{provider}: {usage.total_tokens}" for provider, usage in self.usage_by_provider.items()]
)
- return f"ModelTokenUsage(model={self.model_name}, total={self.total_usage.total_tokens}, backends=[{backends}])"
+ return f"ModelTokenUsage(model={self.model_name}, total={self.total_usage.total_tokens}, providers=[{providers}])"
class TokenUsageTracker:
@@ -123,20 +123,20 @@ class TokenUsageTracker:
self._usage_by_model: Dict[str, ModelTokenUsage] = {}
def update(
- self, model_name: str, backend: str, prompt_tokens: int, completion_tokens: int
+ self, model_name: str, provider: str, prompt_tokens: int, completion_tokens: int
) -> None:
"""Update token usage
Args:
model_name: Model name
- backend: Backend name
+ provider: Provider name
prompt_tokens: Number of input tokens
completion_tokens: Number of output tokens
"""
if model_name not in self._usage_by_model:
self._usage_by_model[model_name] = ModelTokenUsage(model_name)
- self._usage_by_model[model_name].update(backend, prompt_tokens, completion_tokens)
+ self._usage_by_model[model_name].update(provider, prompt_tokens, completion_tokens)
def get_model_usage(self, model_name: str) -> Optional[ModelTokenUsage]:
"""Get token usage for specified model
diff --git a/openviking/storage/observers/vlm_observer.py b/openviking/storage/observers/vlm_observer.py
index 6600648ea..28dc709e5 100644
--- a/openviking/storage/observers/vlm_observer.py
+++ b/openviking/storage/observers/vlm_observer.py
@@ -58,20 +58,20 @@ class VLMObserver(BaseObserver):
total_all = 0
for model_name, model_data in usage_data["usage_by_model"].items():
- for backend_name, backend_data in model_data["usage_by_backend"].items():
+ for provider_name, provider_data in model_data["usage_by_provider"].items():
data.append(
{
"Model": model_name,
- "Backend": backend_name,
- "Prompt": backend_data["prompt_tokens"],
- "Completion": backend_data["completion_tokens"],
- "Total": backend_data["total_tokens"],
- "Last Updated": backend_data["last_updated"],
+ "Provider": provider_name,
+ "Prompt": provider_data["prompt_tokens"],
+ "Completion": provider_data["completion_tokens"],
+ "Total": provider_data["total_tokens"],
+ "Last Updated": provider_data["last_updated"],
}
)
- total_prompt += backend_data["prompt_tokens"]
- total_completion += backend_data["completion_tokens"]
- total_all += backend_data["total_tokens"]
+ total_prompt += provider_data["prompt_tokens"]
+ total_completion += provider_data["completion_tokens"]
+ total_all += provider_data["total_tokens"]
if not data:
return "No token usage data available."
@@ -81,7 +81,7 @@ class VLMObserver(BaseObserver):
# Add total row
total_row = {
"Model": "TOTAL",
- "Backend": "",
+ "Provider": "",
"Prompt": total_prompt,
"Completion": total_completion,
"Total": total_all,
@@ -93,7 +93,7 @@ class VLMObserver(BaseObserver):
index=False,
col_space={
"Model": 30,
- "Backend": 12,
+ "Provider": 12,
"Prompt": 12,
"Completion": 12,
"Total": 12,
diff --git a/openviking/utils/config/embedding_config.py b/openviking/utils/config/embedding_config.py
index f9310575a..5af1861bb 100644
--- a/openviking/utils/config/embedding_config.py
+++ b/openviking/utils/config/embedding_config.py
@@ -14,7 +14,7 @@ class EmbeddingModelConfig(BaseModel):
dimension: Optional[int] = Field(default=None, description="Embedding dimension")
batch_size: int = Field(default=32, description="Batch size for embedding generation")
input: str = Field(default="multimodal", description="Input type: 'text' or 'multimodal'")
- backend: Optional[str] = Field(default="volcengine", description="Backend type: 'openai', 'volcengine', 'vikingdb'")
+ provider: Optional[str] = Field(default="volcengine", description="Provider type: 'openai', 'volcengine', 'vikingdb'")
version: Optional[str] = Field(default=None, description="Model version")
ak: Optional[str] = Field(default=None, description="Access Key ID for VikingDB API")
sk: Optional[str] = Field(default=None, description="Access Key Secretfor VikingDB API")
@@ -27,24 +27,24 @@ class EmbeddingModelConfig(BaseModel):
if not self.model:
raise ValueError("Embedding model name is required")
- if not self.backend:
- raise ValueError("Embedding backend is required")
+ if not self.provider:
+ raise ValueError("Embedding provider is required")
- if self.backend not in ["openai", "volcengine", "vikingdb"]:
+ if self.provider not in ["openai", "volcengine", "vikingdb"]:
raise ValueError(
- f"Invalid embedding backend: '{self.backend}'. Must be one of: 'openai', 'volcengine', 'vikingdb'"
+ f"Invalid embedding provider: '{self.provider}'. Must be one of: 'openai', 'volcengine', 'vikingdb'"
)
- # Backend-specific validation
- if self.backend == "openai":
+ # Provider-specific validation
+ if self.provider == "openai":
if not self.api_key:
- raise ValueError("OpenAI backend requires 'api_key' to be set")
+ raise ValueError("OpenAI provider requires 'api_key' to be set")
- elif self.backend == "volcengine":
+ elif self.provider == "volcengine":
if not self.api_key:
- raise ValueError("Volcengine backend requires 'api_key' to be set")
+ raise ValueError("Volcengine provider requires 'api_key' to be set")
- elif self.backend == "vikingdb":
+ elif self.provider == "vikingdb":
missing = []
if not self.ak:
missing.append("ak")
@@ -55,7 +55,7 @@ class EmbeddingModelConfig(BaseModel):
if missing:
raise ValueError(
- f"VikingDB backend requires the following fields: {', '.join(missing)}"
+ f"VikingDB provider requires the following fields: {', '.join(missing)}"
)
return self
@@ -92,7 +92,7 @@ class EmbeddingConfig(BaseModel):
env_fields = {
"model": str, "api_key": str, "api_base": str,
"dimension": int, "batch_size": int,
- "input": str, "backend": str, "version": str,
+ "input": str, "provider": str, "version": str,
"ak": str, "sk": str, "region": str, "host": str
}
@@ -127,11 +127,11 @@ class EmbeddingConfig(BaseModel):
)
return self
- def _create_embedder(self, backend: str, embedder_type: str, config: EmbeddingModelConfig):
- """Factory method to create embedder instance based on backend and type.
+ def _create_embedder(self, provider: str, embedder_type: str, config: EmbeddingModelConfig):
+ """Factory method to create embedder instance based on provider and type.
Args:
- backend: Backend type ('openai', 'volcengine', 'vikingdb')
+ provider: Provider type ('openai', 'volcengine', 'vikingdb')
embedder_type: Embedder type ('dense', 'sparse', 'hybrid')
config: EmbeddingModelConfig instance
@@ -139,7 +139,7 @@ class EmbeddingConfig(BaseModel):
Embedder instance
Raises:
- ValueError: If backend/type combination is not supported
+ ValueError: If provider/type combination is not supported
"""
from openviking.models.embedder import (
OpenAIDenseEmbedder,
@@ -151,7 +151,7 @@ class EmbeddingConfig(BaseModel):
VikingDBHybridEmbedder
)
- # Factory registry: (backend, type) -> (embedder_class, param_builder)
+ # Factory registry: (provider, type) -> (embedder_class, param_builder)
factory_registry = {
('openai', 'dense'): (
OpenAIDenseEmbedder,
@@ -229,10 +229,10 @@ class EmbeddingConfig(BaseModel):
),
}
- key = (backend, embedder_type)
+ key = (provider, embedder_type)
if key not in factory_registry:
raise ValueError(
- f"Unsupported combination: backend='{backend}', type='{embedder_type}'. "
+ f"Unsupported combination: provider='{provider}', type='{embedder_type}'. "
f"Supported combinations: {list(factory_registry.keys())}"
)
@@ -252,15 +252,15 @@ class EmbeddingConfig(BaseModel):
from openviking.models.embedder import CompositeHybridEmbedder
if self.hybrid:
- return self._create_embedder(self.hybrid.backend.lower(), 'hybrid', self.hybrid)
+ return self._create_embedder(self.hybrid.provider.lower(), 'hybrid', self.hybrid)
if self.dense and self.sparse:
- dense_embedder = self._create_embedder(self.dense.backend.lower(), 'dense', self.dense)
- sparse_embedder = self._create_embedder(self.sparse.backend.lower(), 'sparse', self.sparse)
+ dense_embedder = self._create_embedder(self.dense.provider.lower(), 'dense', self.dense)
+ sparse_embedder = self._create_embedder(self.sparse.provider.lower(), 'sparse', self.sparse)
return CompositeHybridEmbedder(dense_embedder, sparse_embedder)
if self.dense:
- return self._create_embedder(self.dense.backend.lower(), 'dense', self.dense)
+ return self._create_embedder(self.dense.provider.lower(), 'dense', self.dense)
raise ValueError("No embedding configuration found (dense, sparse, or hybrid)")
diff --git a/openviking/utils/config/vlm_config.py b/openviking/utils/config/vlm_config.py
index ec61a7a59..dfdb30765 100644
--- a/openviking/utils/config/vlm_config.py
+++ b/openviking/utils/config/vlm_config.py
@@ -9,15 +9,15 @@ from openviking.models.vlm import VLMFactory, VLMBase
class VLMConfig(BaseModel):
- """VLM configuration, supports multiple backends (openai, volcengine)."""
+ """VLM configuration, supports multiple providers (openai, volcengine)."""
model: Optional[str] = Field(default=None, description="Model name")
api_key: Optional[str] = Field(default=None, description="API key")
api_base: Optional[str] = Field(default=None, description="API base URL")
temperature: float = Field(default=0.0, description="Generation temperature")
max_retries: int = Field(default=2, description="Maximum retry attempts")
- backend: Literal["openai", "volcengine"] = Field(
- default="openai", description="Backend provider"
+ provider: Literal["openai", "volcengine"] = Field(
+ default="volcengine", description="Provider type"
)
_vlm_instance: Optional[VLMBase] = None
@@ -34,13 +34,14 @@ class VLMConfig(BaseModel):
"api_key": "OPENVIKING_VLM_API_KEY",
"model": "OPENVIKING_VLM_MODEL",
"api_base": "OPENVIKING_VLM_API_BASE",
- "backend": "OPENVIKING_VLM_BACKEND",
+ "provider": "OPENVIKING_VLM_PROVIDER",
}
for field, env_var in env_mapping.items():
if data.get(field) is None:
env_val = os.getenv(env_var)
if env_val is not None:
data[field] = env_val
+
return data
@model_validator(mode='after')
diff --git a/tests/misc/test_config_validation.py b/tests/misc/test_config_validation.py
index eff4f87d6..4e833c431 100644
--- a/tests/misc/test_config_validation.py
+++ b/tests/misc/test_config_validation.py
@@ -104,12 +104,12 @@ def test_embedding_validation():
except ValueError as e:
print(f" Correctly raised exception: {e}")
- # Test 2: OpenAI backend missing api_key
- print("\n2. Test OpenAI backend missing api_key...")
+ # Test 2: OpenAI provider missing api_key
+ print("\n2. Test OpenAI provider missing api_key...")
try:
config = EmbeddingConfig(
dense=EmbeddingModelConfig(
- backend="openai",
+ provider="openai",
model="text-embedding-3-small"
)
)
@@ -117,12 +117,12 @@ def test_embedding_validation():
except ValueError as e:
print(f" Correctly raised exception: {e}")
- # Test 3: OpenAI backend complete config
- print("\n3. Test OpenAI backend complete config...")
+ # Test 3: OpenAI provider complete config
+ print("\n3. Test OpenAI provider complete config...")
try:
config = EmbeddingConfig(
dense=EmbeddingModelConfig(
- backend="openai",
+ provider="openai",
model="text-embedding-3-small",
api_key="fake-api-key-for-testing",
dimension=1536
@@ -160,7 +160,7 @@ def test_vlm_validation():
config = VLMConfig(
model="gpt-4",
api_key="fake-api-key-for-testing",
- backend="openai"
+ provider="openai"
)
print(f" Pass")
except ValueError as e: