Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Python agent-contextagent-memoryagent-pluginsagentic-ragcontext-databasedsh-pluginknowledge-basememorymemory-managementself-evolvingvector-database
Revert "fix: use provider instead of backend"
qin-ctxb5d8149dd92a
diff --git a/README.md b/README.mdindex dd8ab82d5..6b92629b1 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- "provider" : "<provider-type>", // Provider type (volcengine or openai)+ "backend" : "<backend-type>", // Backend 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- "provider" : "<provider-type>", // Provider type (volcengine or openai)+ "backend" : "<backend-type>", // Backend 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",- "provider" : "volcengine",+ "backend" : "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",- "provider" : "volcengine",+ "backend" : "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",- "provider" : "openai",+ "backend" : "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",- "provider" : "openai",+ "backend" : "openai","model" : "gpt-4-vision-preview"}}diff --git a/README_CN.md b/README_CN.mdindex 2167ac68a..ad550a391 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- "provider" : "<provider-type>", // 提供商类型(volcengine 或 openai)+ "backend" : "<backend-type>", // 后端类型(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- "provider" : "<provider-type>", // 提供商类型(volcengine 或 openai)+ "backend" : "<backend-type>", // 后端类型(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",- "provider" : "volcengine",+ "backend" : "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",- "provider" : "volcengine",+ "backend" : "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",- "provider" : "openai",+ "backend" : "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",- "provider" : "openai",+ "backend" : "openai","model" : "gpt-4-vision-preview"}}diff --git a/docs/en/configuration/configuration.md b/docs/en/configuration/configuration.mdindex 45ba3942b..2150dbfd5 100644--- a/docs/en/configuration/configuration.md+++ b/docs/en/configuration/configuration.md@@ -10,14 +10,13 @@ Create `ov.conf` in your project directory:{"embedding": {"dense": {- "provider": "volcengine",+ "backend": "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"},@@ -49,7 +48,7 @@ Embedding model configuration for vector search.{"embedding": {"dense": {- "provider": "volcengine",+ "backend": "volcengine","api_key": "your-api-key","model": "doubao-embedding-vision-250615","dimension": 1024,@@ -68,7 +67,6 @@ 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"@@ -154,7 +152,7 @@ config = OpenVikingConfig(),embedding=EmbeddingConfig(dense=DenseEmbeddingConfig(- provider="volcengine",+ backend="volcengine",api_key="your-api-key",model="doubao-embedding-vision-250615",dimension=1024@@ -173,7 +171,7 @@ client = ov.AsyncOpenViking(config=config){"embedding": {"dense": {- "provider": "volcengine",+ "backend": "volcengine","api_key": "string","model": "string","dimension": 1024,@@ -181,7 +179,6 @@ 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.mdindex da58d9865..b03e20f40 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": {- "provider": "volcengine",+ "backend": "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 ||-----------|------|-------------|-| `provider` | str | `"volcengine"` |+| `backend` | 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": {- "provider": "volcengine",+ "backend": "volcengine","model": "doubao-embedding-vision-250615","dimension": 1024}@@ -69,7 +69,7 @@ from openviking.utils.config import EmbeddingConfig, DenseEmbeddingConfigembedding_config = EmbeddingConfig(dense=DenseEmbeddingConfig(- provider="volcengine",+ backend="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.mdindex 3fbe0fcb8..df7c10aa3 100644--- a/docs/en/configuration/llm.md+++ b/docs/en/configuration/llm.md@@ -9,7 +9,6 @@ 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.mdindex 87b1aec80..cc26e5eb9 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": {- "provider": "volcengine",+ "backend": "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": {- "provider": "volcengine",+ "backend": "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": {- "provider": "volcengine",+ "backend": "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 provider:**+**支持的 backend:**- `openai`: OpenAI Embedding API- `volcengine`: 火山引擎 Embedding API- `vikingdb`: VikingDB Embedding API-**vikingdb provider 配置示例:**+**vikingdb backend 配置示例:**```json{"embedding": {"dense": {- "provider": "vikingdb",+ "backend": "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": {- "provider": "volcengine",+ "backend": "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": {- "provider": "volcengine",+ "backend": "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": {- "provider": "volcengine",+ "backend": "volcengine","api_key": "your-api-key","model": "doubao-embedding-vision-250615","dimension": 1024},"sparse": {- "provider": "volcengine",+ "backend": "volcengine","api_key": "your-api-key","model": "bm25-sparse-v1"}@@ -154,7 +154,6 @@ 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"@@ -240,7 +239,7 @@ config = OpenVikingConfig(),embedding=EmbeddingConfig(dense=DenseEmbeddingConfig(- provider="volcengine",+ backend="volcengine",api_key="your-api-key",model="doubao-embedding-vision-250615",dimension=1024@@ -259,7 +258,7 @@ client = ov.AsyncOpenViking(config=config){"embedding": {"dense": {- "provider": "volcengine",+ "backend": "volcengine","api_key": "string","model": "string","dimension": 1024,@@ -267,7 +266,6 @@ 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.mdindex 626ee5cbe..49dd881c8 100644--- a/docs/zh/configuration/embedding.md+++ b/docs/zh/configuration/embedding.md@@ -8,7 +8,7 @@{"embedding": {"dense": {- "provider": "volcengine",+ "backend": "volcengine","api_key": "your-volcengine-api-key","model": "doubao-embedding-vision-250615","dimension": 1024,@@ -22,7 +22,7 @@| 参数 | 类型 | 说明 ||------|------|------|-| `provider` | str | `"volcengine"` |+| `backend` | 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": {- "provider": "volcengine",+ "backend": "volcengine","model": "doubao-embedding-vision-250615","dimension": 1024}@@ -69,7 +69,7 @@ from openviking.utils.config import EmbeddingConfig, DenseEmbeddingConfigembedding_config = EmbeddingConfig(dense=DenseEmbeddingConfig(- provider="volcengine",+ backend="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.mdindex a8a714ddf..ca8047d93 100644--- a/docs/zh/configuration/llm.md+++ b/docs/zh/configuration/llm.md@@ -9,7 +9,6 @@```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.exampleindex 353d94e73..b853b6ad9 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,- "provider": "volcengine",+ "backend": "volcengine","input": "multimodal"}},@@ -39,7 +39,7 @@"api_base": "https://ark.cn-beijing.volces.com/api/v3","temperature": 0.0,"max_retries": 2,- "provider": "volcengine"+ "backend": "volcengine"},"rerank": {"ak": null,diff --git a/openviking/models/vlm/backends/openai_vlm.py b/openviking/models/vlm/backends/openai_vlm.pyindex 27b014498..50054219f 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 = Noneself._async_client = None- # Ensure provider type is correct- self.provider = "openai"+ # Ensure backend type is correct+ self.backend = "openai"def get_client(self):"""Get sync client"""@@ -46,7 +46,7 @@ class OpenAIVLM(VLMBase):completion_tokens = response.usage.completion_tokensself.update_token_usage(model_name=self.model or "gpt-4o-mini",- provider=self.provider,+ backend="openai",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.pyindex 8f1d2e97c..3ee951707 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 = Noneself._async_client = None- # Ensure provider type is correct- self.provider = "volcengine"+ # Ensure backend type is correct+ self.backend = "volcengine"# VolcEngine-specific defaultsif not self.api_base:diff --git a/openviking/models/vlm/base.py b/openviking/models/vlm/base.pyindex 7111fe162..fc1c797e7 100644--- a/openviking/models/vlm/base.py+++ b/openviking/models/vlm/base.py@@ -14,8 +14,7 @@ class VLMBase(ABC):def __init__(self, config: Dict[str, Any]):self.config = config- self.provider = config.get("backend", "openai")-+ self.backend = config.get("backend", "openai")self.model = config.get("model")self.api_key = config.get("api_key")self.api_base = config.get("api_base")@@ -59,19 +58,19 @@ class VLMBase(ABC):# Token usage tracking methodsdef update_token_usage(- self, model_name: str, provider: str, prompt_tokens: int, completion_tokens: int+ self, model_name: str, backend: str, prompt_tokens: int, completion_tokens: int) -> None:"""Update token usageArgs:model_name: Model name- provider: Provider name (openai, volcengine)+ backend: Backend name (openai, volcengine)prompt_tokens: Number of prompt tokenscompletion_tokens: Number of completion tokens"""self._token_tracker.update(model_name=model_name,- provider=provider,+ backend=backend,prompt_tokens=prompt_tokens,completion_tokens=completion_tokens,)@@ -111,27 +110,26 @@ class VLMFactory:"""Create VLM instanceArgs:- config: VLM config, must contain 'provider' field+ config: VLM config, must contain 'backend' fieldReturns:VLMBase: VLM instanceRaises:- ValueError: If provider is not supported+ ValueError: If backend is not supportedImportError: If related dependencies are not installed"""- provider = config.get("backend", "openai")-- if provider == "openai":+ backend = config.get("backend", "openai")+ if backend == "openai":from .backends.openai_vlm import OpenAIVLMreturn OpenAIVLM(config)- elif provider == "volcengine":+ elif backend == "volcengine":from .backends.volcengine_vlm import VolcEngineVLMreturn VolcEngineVLM(config)else:- raise ValueError(f"Unsupported VLM provider: {provider}")+ raise ValueError(f"Unsupported VLM backend: {backend}")@staticmethoddef get_available_backends() -> List[str]:- """Get list of available providers"""+ """Get list of available backends"""return ["openai", "volcengine"]diff --git a/openviking/models/vlm/token_usage.py b/openviking/models/vlm/token_usage.pyindex f1b511946..6b29f635d 100644--- a/openviking/models/vlm/token_usage.py+++ b/openviking/models/vlm/token_usage.py@@ -62,35 +62,35 @@ class ModelTokenUsage:model_name: strtotal_usage: TokenUsage = field(default_factory=TokenUsage)- usage_by_provider: Dict[str, TokenUsage] = field(default_factory=dict)+ usage_by_backend: Dict[str, TokenUsage] = field(default_factory=dict)- def update(self, provider: str, prompt_tokens: int, completion_tokens: int) -> None:- """Update token usage for specified provider+ def update(self, backend: str, prompt_tokens: int, completion_tokens: int) -> None:+ """Update token usage for specified backendArgs:- provider: Provider name (openai, volcengine)+ backend: Backend name (openai, volcengine)prompt_tokens: Number of input tokenscompletion_tokens: Number of output tokens"""# Update total usageself.total_usage.update(prompt_tokens, completion_tokens)- # Update provider usage- if provider not in self.usage_by_provider:- self.usage_by_provider[provider] = TokenUsage()+ # Update backend usage+ if backend not in self.usage_by_backend:+ self.usage_by_backend[backend] = TokenUsage()- self.usage_by_provider[provider].update(prompt_tokens, completion_tokens)+ self.usage_by_backend[backend].update(prompt_tokens, completion_tokens)- def get_provider_usage(self, provider: str) -> Optional[TokenUsage]:- """Get token usage for specified provider+ def get_backend_usage(self, backend: str) -> Optional[TokenUsage]:+ """Get token usage for specified backendArgs:- provider: Provider name+ backend: Backend nameReturns:- TokenUsage object, or None if provider doesn't exist+ TokenUsage object, or None if backend doesn't exist"""- return self.usage_by_provider.get(provider)+ return self.usage_by_backend.get(backend)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_provider": {},+ "usage_by_backend": {},}- for provider, usage in self.usage_by_provider.items():- result["usage_by_provider"][provider] = usage.to_dict()+ for backend, usage in self.usage_by_backend.items():+ result["usage_by_backend"][backend] = usage.to_dict()return resultdef __str__(self) -> str:- providers = ", ".join(- [f"{provider}: {usage.total_tokens}" for provider, usage in self.usage_by_provider.items()]+ backends = ", ".join(+ [f"{backend}: {usage.total_tokens}" for backend, usage in self.usage_by_backend.items()])- return f"ModelTokenUsage(model={self.model_name}, total={self.total_usage.total_tokens}, providers=[{providers}])"+ return f"ModelTokenUsage(model={self.model_name}, total={self.total_usage.total_tokens}, backends=[{backends}])"class TokenUsageTracker:@@ -123,20 +123,20 @@ class TokenUsageTracker:self._usage_by_model: Dict[str, ModelTokenUsage] = {}def update(- self, model_name: str, provider: str, prompt_tokens: int, completion_tokens: int+ self, model_name: str, backend: str, prompt_tokens: int, completion_tokens: int) -> None:"""Update token usageArgs:model_name: Model name- provider: Provider name+ backend: Backend nameprompt_tokens: Number of input tokenscompletion_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(provider, prompt_tokens, completion_tokens)+ self._usage_by_model[model_name].update(backend, prompt_tokens, completion_tokens)def get_model_usage(self, model_name: str) -> Optional[ModelTokenUsage]:"""Get token usage for specified modeldiff --git a/openviking/storage/observers/vlm_observer.py b/openviking/storage/observers/vlm_observer.pyindex 28dc709e5..6600648ea 100644--- a/openviking/storage/observers/vlm_observer.py+++ b/openviking/storage/observers/vlm_observer.py@@ -58,20 +58,20 @@ class VLMObserver(BaseObserver):total_all = 0for model_name, model_data in usage_data["usage_by_model"].items():- for provider_name, provider_data in model_data["usage_by_provider"].items():+ for backend_name, backend_data in model_data["usage_by_backend"].items():data.append({"Model": model_name,- "Provider": provider_name,- "Prompt": provider_data["prompt_tokens"],- "Completion": provider_data["completion_tokens"],- "Total": provider_data["total_tokens"],- "Last Updated": provider_data["last_updated"],+ "Backend": backend_name,+ "Prompt": backend_data["prompt_tokens"],+ "Completion": backend_data["completion_tokens"],+ "Total": backend_data["total_tokens"],+ "Last Updated": backend_data["last_updated"],})- total_prompt += provider_data["prompt_tokens"]- total_completion += provider_data["completion_tokens"]- total_all += provider_data["total_tokens"]+ total_prompt += backend_data["prompt_tokens"]+ total_completion += backend_data["completion_tokens"]+ total_all += backend_data["total_tokens"]if not data:return "No token usage data available."@@ -81,7 +81,7 @@ class VLMObserver(BaseObserver):# Add total rowtotal_row = {"Model": "TOTAL",- "Provider": "",+ "Backend": "","Prompt": total_prompt,"Completion": total_completion,"Total": total_all,@@ -93,7 +93,7 @@ class VLMObserver(BaseObserver):index=False,col_space={"Model": 30,- "Provider": 12,+ "Backend": 12,"Prompt": 12,"Completion": 12,"Total": 12,diff --git a/openviking/utils/config/embedding_config.py b/openviking/utils/config/embedding_config.pyindex 5af1861bb..f9310575a 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'")- provider: Optional[str] = Field(default="volcengine", description="Provider type: 'openai', 'volcengine', 'vikingdb'")+ backend: Optional[str] = Field(default="volcengine", description="Backend 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.provider:- raise ValueError("Embedding provider is required")+ if not self.backend:+ raise ValueError("Embedding backend is required")- if self.provider not in ["openai", "volcengine", "vikingdb"]:+ if self.backend not in ["openai", "volcengine", "vikingdb"]:raise ValueError(- f"Invalid embedding provider: '{self.provider}'. Must be one of: 'openai', 'volcengine', 'vikingdb'"+ f"Invalid embedding backend: '{self.backend}'. Must be one of: 'openai', 'volcengine', 'vikingdb'")- # Provider-specific validation- if self.provider == "openai":+ # Backend-specific validation+ if self.backend == "openai":if not self.api_key:- raise ValueError("OpenAI provider requires 'api_key' to be set")+ raise ValueError("OpenAI backend requires 'api_key' to be set")- elif self.provider == "volcengine":+ elif self.backend == "volcengine":if not self.api_key:- raise ValueError("Volcengine provider requires 'api_key' to be set")+ raise ValueError("Volcengine backend requires 'api_key' to be set")- elif self.provider == "vikingdb":+ elif self.backend == "vikingdb":missing = []if not self.ak:missing.append("ak")@@ -55,7 +55,7 @@ class EmbeddingModelConfig(BaseModel):if missing:raise ValueError(- f"VikingDB provider requires the following fields: {', '.join(missing)}"+ f"VikingDB backend 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, "provider": str, "version": str,+ "input": str, "backend": str, "version": str,"ak": str, "sk": str, "region": str, "host": str}@@ -127,11 +127,11 @@ class EmbeddingConfig(BaseModel):)return self- def _create_embedder(self, provider: str, embedder_type: str, config: EmbeddingModelConfig):- """Factory method to create embedder instance based on provider and type.+ def _create_embedder(self, backend: str, embedder_type: str, config: EmbeddingModelConfig):+ """Factory method to create embedder instance based on backend and type.Args:- provider: Provider type ('openai', 'volcengine', 'vikingdb')+ backend: Backend type ('openai', 'volcengine', 'vikingdb')embedder_type: Embedder type ('dense', 'sparse', 'hybrid')config: EmbeddingModelConfig instance@@ -139,7 +139,7 @@ class EmbeddingConfig(BaseModel):Embedder instanceRaises:- ValueError: If provider/type combination is not supported+ ValueError: If backend/type combination is not supported"""from openviking.models.embedder import (OpenAIDenseEmbedder,@@ -151,7 +151,7 @@ class EmbeddingConfig(BaseModel):VikingDBHybridEmbedder)- # Factory registry: (provider, type) -> (embedder_class, param_builder)+ # Factory registry: (backend, type) -> (embedder_class, param_builder)factory_registry = {('openai', 'dense'): (OpenAIDenseEmbedder,@@ -229,10 +229,10 @@ class EmbeddingConfig(BaseModel):),}- key = (provider, embedder_type)+ key = (backend, embedder_type)if key not in factory_registry:raise ValueError(- f"Unsupported combination: provider='{provider}', type='{embedder_type}'. "+ f"Unsupported combination: backend='{backend}', type='{embedder_type}'. "f"Supported combinations: {list(factory_registry.keys())}")@@ -252,15 +252,15 @@ class EmbeddingConfig(BaseModel):from openviking.models.embedder import CompositeHybridEmbedderif self.hybrid:- return self._create_embedder(self.hybrid.provider.lower(), 'hybrid', self.hybrid)+ return self._create_embedder(self.hybrid.backend.lower(), 'hybrid', self.hybrid)if self.dense and 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)+ dense_embedder = self._create_embedder(self.dense.backend.lower(), 'dense', self.dense)+ sparse_embedder = self._create_embedder(self.sparse.backend.lower(), 'sparse', self.sparse)return CompositeHybridEmbedder(dense_embedder, sparse_embedder)if self.dense:- return self._create_embedder(self.dense.provider.lower(), 'dense', self.dense)+ return self._create_embedder(self.dense.backend.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.pyindex dfdb30765..ec61a7a59 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, VLMBaseclass VLMConfig(BaseModel):- """VLM configuration, supports multiple providers (openai, volcengine)."""+ """VLM configuration, supports multiple backends (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")- provider: Literal["openai", "volcengine"] = Field(- default="volcengine", description="Provider type"+ backend: Literal["openai", "volcengine"] = Field(+ default="openai", description="Backend provider")_vlm_instance: Optional[VLMBase] = None@@ -34,14 +34,13 @@ class VLMConfig(BaseModel):"api_key": "OPENVIKING_VLM_API_KEY","model": "OPENVIKING_VLM_MODEL","api_base": "OPENVIKING_VLM_API_BASE",- "provider": "OPENVIKING_VLM_PROVIDER",+ "backend": "OPENVIKING_VLM_BACKEND",}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.pyindex 4e833c431..eff4f87d6 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 provider missing api_key- print("\n2. Test OpenAI provider missing api_key...")+ # Test 2: OpenAI backend missing api_key+ print("\n2. Test OpenAI backend missing api_key...")try:config = EmbeddingConfig(dense=EmbeddingModelConfig(- provider="openai",+ backend="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 provider complete config- print("\n3. Test OpenAI provider complete config...")+ # Test 3: OpenAI backend complete config+ print("\n3. Test OpenAI backend complete config...")try:config = EmbeddingConfig(dense=EmbeddingModelConfig(- provider="openai",+ backend="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",- provider="openai"+ backend="openai")print(f" Pass")except ValueError as e: