Feat/rerank model (#2137)

* feat(provider): add rerank model management as a core model type * feat(provider): add rerank support to existing requesters and new rerank providers * feat(web): add rerank model management UI and pipeline config * fix(provider): correct rerank support_type after verification - Add rerank to OpenRouter (confirmed /api/v1/rerank endpoint) - Remove rerank from Ollama (no native support, PR #7219 unmerged) - Remove rerank from JiekouAI (no rerank docs found, URL path mismatch) * fix(provider): remove alru_cache from model getters and add rerank param hints * fix: resolve lint errors - Remove unused alru_cache import from modelmgr.py - Remove unused error_message variable in invoke_rerank - Fix prettier formatting in frontend files Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix: remove unused exception variable - Change `except Exception as e:` to `except Exception:` since e is not used - Fix prettier formatting in ProviderCard.tsx Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix: apply ruff format Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * feat(template): add rerank config fields to default pipeline config Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * chore: remove PR.md Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> * fix(ui): remove duplicate rerank model form in AddModelPopover The form was being rendered twice: once in TabsContent manual mode and again in a separate conditional block for rerank tab. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
2026-07-16 17:36:07 +00:00 · 2026-04-20 23:32:36 +08:00
parent 5a5c4295b1
commit 323481d69b
41 changed files with 950 additions and 41 deletions
@@ -9,7 +9,6 @@ from ...discover import engine
 from . import token
 from ...entity.persistence import model as persistence_model
 from ...entity.errors import provider as provider_errors
-from async_lru import alru_cache


 class ModelManager:
@@ -24,6 +23,8 @@ class ModelManager:

    embedding_models: list[requester.RuntimeEmbeddingModel]

+    rerank_models: list[requester.RuntimeRerankModel]
+
    requester_components: list[engine.Component]

    requester_dict: dict[str, type[requester.ProviderAPIRequester]]
@@ -32,6 +33,7 @@ class ModelManager:
        self.ap = ap
        self.llm_models = []
        self.embedding_models = []
+        self.rerank_models = []
        self.requester_components = []
        self.requester_dict = {}

@@ -64,8 +66,7 @@ class ModelManager:

        self.llm_models = []
        self.embedding_models = []
-
-        # Load all providers first
+        self.rerank_models = []
        self.provider_dict = {}
        providers_result = await self.ap.persistence_mgr.execute_async(
            sqlalchemy.select(persistence_model.ModelProvider)
@@ -110,6 +111,22 @@ class ModelManager:
            except Exception as e:
                self.ap.logger.error(f'Failed to load model {embedding_model.uuid}: {e}\n{traceback.format_exc()}')

+        # Load rerank models
+        result = await self.ap.persistence_mgr.execute_async(sqlalchemy.select(persistence_model.RerankModel))
+        rerank_models = result.all()
+        for rerank_model in rerank_models:
+            try:
+                provider = self.provider_dict.get(rerank_model.provider_uuid)
+                if provider is None:
+                    self.ap.logger.warning(
+                        f'Provider {rerank_model.provider_uuid} not found for model {rerank_model.uuid}'
+                    )
+                    continue
+                runtime_rerank_model = await self.load_rerank_model_with_provider(rerank_model, provider)
+                self.rerank_models.append(runtime_rerank_model)
+            except Exception as e:
+                self.ap.logger.error(f'Failed to load model {rerank_model.uuid}: {e}\n{traceback.format_exc()}')
+
    async def sync_new_models_from_space(self):
        """Sync models from Space"""
        space_model_provider = await self.ap.persistence_mgr.execute_async(
@@ -212,6 +229,26 @@ class ModelManager:

        return runtime_embedding_model

+    async def init_temporary_runtime_rerank_model(
+        self,
+        model_info: dict,
+    ) -> requester.RuntimeRerankModel:
+        """Initialize runtime rerank model from dict (for testing)"""
+        provider_info = model_info.get('provider', {})
+        runtime_provider = await self.load_provider(provider_info)
+
+        runtime_rerank_model = requester.RuntimeRerankModel(
+            model_entity=persistence_model.RerankModel(
+                uuid=model_info.get('uuid', ''),
+                name=model_info.get('name', ''),
+                provider_uuid='',
+                extra_args=model_info.get('extra_args', {}),
+            ),
+            provider=runtime_provider,
+        )
+
+        return runtime_rerank_model
+
    async def load_provider(
        self, provider_info: persistence_model.ModelProvider | sqlalchemy.Row | dict
    ) -> requester.RuntimeProvider:
@@ -269,6 +306,9 @@ class ModelManager:
        for model in self.embedding_models:
            if model.provider.provider_entity.uuid == provider_uuid:
                model.provider = new_runtime_provider
+        for model in self.rerank_models:
+            if model.provider.provider_entity.uuid == provider_uuid:
+                model.provider = new_runtime_provider

        # update ref in provider dict
        self.provider_dict[provider_uuid] = new_runtime_provider
@@ -305,6 +345,22 @@ class ModelManager:

        return runtime_embedding_model

+    async def load_rerank_model_with_provider(
+        self,
+        model_info: persistence_model.RerankModel | sqlalchemy.Row,
+        provider: requester.RuntimeProvider,
+    ) -> requester.RuntimeRerankModel:
+        """Load rerank model with provider info"""
+        if isinstance(model_info, sqlalchemy.Row):
+            model_info = persistence_model.RerankModel(**model_info._mapping)
+
+        runtime_rerank_model = requester.RuntimeRerankModel(
+            model_entity=model_info,
+            provider=provider,
+        )
+
+        return runtime_rerank_model
+
    async def load_llm_model(self, model_info: dict):
        """Load LLM model from dict (with provider info)"""
        provider_info = model_info.get('provider', {})
@@ -352,7 +408,6 @@ class ModelManager:

        await self.load_embedding_model_with_provider(model_entity, provider_entity)

-    @alru_cache(ttl=60 * 5)
    async def get_model_by_uuid(self, uuid: str) -> requester.RuntimeLLMModel:
        """Get LLM model by uuid"""
        for model in self.llm_models:
@@ -360,7 +415,6 @@ class ModelManager:
                return model
        raise ValueError(f'LLM model {uuid} not found')

-    @alru_cache(ttl=60 * 5)
    async def get_embedding_model_by_uuid(self, uuid: str) -> requester.RuntimeEmbeddingModel:
        """Get embedding model by uuid"""
        for model in self.embedding_models:
@@ -368,6 +422,13 @@ class ModelManager:
                return model
        raise ValueError(f'Embedding model {uuid} not found')

+    async def get_rerank_model_by_uuid(self, uuid: str) -> requester.RuntimeRerankModel:
+        """Get rerank model by uuid"""
+        for model in self.rerank_models:
+            if model.model_entity.uuid == uuid:
+                return model
+        raise ValueError(f'Rerank model {uuid} not found')
+
    async def remove_llm_model(self, model_uuid: str):
        """Remove LLM model"""
        for model in self.llm_models:
@@ -382,6 +443,13 @@ class ModelManager:
                self.embedding_models.remove(model)
                return

+    async def remove_rerank_model(self, model_uuid: str):
+        """Remove rerank model"""
+        for model in self.rerank_models:
+            if model.model_entity.uuid == model_uuid:
+                self.rerank_models.remove(model)
+                return
+
    def get_available_requesters_info(self, model_type: str) -> list[dict]:
        """Get all available requesters"""
        if model_type != '':
@@ -247,6 +247,40 @@ class RuntimeProvider:
            except Exception as monitor_err:
                self.requester.ap.logger.error(f'[Monitoring] Failed to record embedding call: {monitor_err}')

+    async def invoke_rerank(
+        self,
+        model: RuntimeRerankModel,
+        query: str,
+        documents: typing.List[str],
+        extra_args: dict[str, typing.Any] = {},
+    ) -> typing.List[dict]:
+        """Bridge method for invoking rerank with monitoring"""
+        start_time = time.time()
+        status = 'success'
+
+        try:
+            result = await self.requester.invoke_rerank(
+                model=model,
+                query=query,
+                documents=documents,
+                extra_args=extra_args,
+            )
+            return result
+
+        except Exception:
+            status = 'error'
+            raise
+        finally:
+            duration_ms = int((time.time() - start_time) * 1000)
+
+            try:
+                self.requester.ap.logger.debug(
+                    f'[Rerank] model={model.model_entity.name} docs={len(documents)} '
+                    f'duration={duration_ms}ms status={status}'
+                )
+            except Exception as monitor_err:
+                self.requester.ap.logger.error(f'[Monitoring] Failed to record rerank call: {monitor_err}')
+

 class RuntimeLLMModel:
    """运行时模型"""
@@ -284,6 +318,24 @@ class RuntimeEmbeddingModel:
        self.provider = provider


+class RuntimeRerankModel:
+    """运行时 Rerank 模型"""
+
+    model_entity: persistence_model.RerankModel
+    """模型数据"""
+
+    provider: RuntimeProvider
+    """提供商实例"""
+
+    def __init__(
+        self,
+        model_entity: persistence_model.RerankModel,
+        provider: RuntimeProvider,
+    ):
+        self.model_entity = model_entity
+        self.provider = provider
+
+
 class ProviderAPIRequester(metaclass=abc.ABCMeta):
    """Provider API请求器"""

@@ -376,3 +428,23 @@ class ProviderAPIRequester(metaclass=abc.ABCMeta):
            或者 tuple[typing.List[typing.List[float]], dict]: 返回 (embedding 向量, usage_info)
        """
        pass
+
+    async def invoke_rerank(
+        self,
+        model: RuntimeRerankModel,
+        query: str,
+        documents: typing.List[str],
+        extra_args: dict[str, typing.Any] = {},
+    ) -> typing.List[dict]:
+        """调用 Rerank API
+
+        Args:
+            model (RuntimeRerankModel): 使用的模型信息
+            query (str): 查询文本
+            documents (typing.List[str]): 待重排序的文档列表
+            extra_args (dict[str, typing.Any], optional): 额外的参数. Defaults to {}.
+
+        Returns:
+            typing.List[dict]: [{"index": int, "relevance_score": float}, ...]
+        """
+        raise NotImplementedError('This requester does not support rerank')
@@ -25,6 +25,7 @@ spec:
  support_type:
  - llm
  - text-embedding
+  - rerank
  provider_category: maas
 execution:
  python:
@@ -24,6 +24,7 @@ spec:
    default: 120
  support_type:
  - llm
+  - rerank
  provider_category: maas
 execution:
  python:
@@ -615,3 +615,88 @@ class OpenAIChatCompletions(requester.ProviderAPIRequester):
            raise errors.RequesterError(f'请求过于频繁或余额不足: {e.message}')
        except openai.APIError as e:
            raise errors.RequesterError(f'请求错误: {e.message}')
+
+    async def invoke_rerank(
+        self,
+        model: requester.RuntimeRerankModel,
+        query: str,
+        documents: typing.List[str],
+        extra_args: dict[str, typing.Any] = {},
+    ) -> typing.List[dict]:
+        """Standard /rerank endpoint (Jina/Cohere/SiliconFlow/Voyage/DashScope compatible)
+
+        Supports extra_args from model.extra_args:
+        - rerank_url: full URL override (e.g. "https://dashscope.aliyuncs.com/compatible-api/v1/reranks")
+        - rerank_path: path override appended to base_url (e.g. "reranks" instead of default "rerank")
+        - Any other fields are merged into the request payload.
+        """
+        api_key = model.provider.token_mgr.get_token()
+        base_url = self.requester_cfg.get('base_url', '').rstrip('/')
+        timeout = self.requester_cfg.get('timeout', 120)
+
+        merged_args = {}
+        if model.model_entity.extra_args:
+            merged_args.update(model.model_entity.extra_args)
+        if extra_args:
+            merged_args.update(extra_args)
+
+        rerank_url = merged_args.pop('rerank_url', None)
+        rerank_path = merged_args.pop('rerank_path', 'rerank')
+        if not rerank_url:
+            rerank_url = f'{base_url}/{rerank_path}'
+
+        headers = {
+            'Content-Type': 'application/json',
+            'Authorization': f'Bearer {api_key}',
+        }
+
+        payload = {
+            'model': model.model_entity.name,
+            'query': query,
+            'documents': documents[:64],
+            'top_n': min(len(documents), 64),
+        }
+
+        if merged_args:
+            payload.update(merged_args)
+
+        try:
+            async with httpx.AsyncClient(trust_env=True, timeout=timeout) as client:
+                resp = await client.post(rerank_url, headers=headers, json=payload)
+                resp.raise_for_status()
+                data = resp.json()
+
+            results = self._parse_rerank_response(data)
+
+            if results:
+                scores = [r.get('relevance_score', 0.0) for r in results]
+                min_score = min(scores)
+                max_score = max(scores)
+                if max_score - min_score > 1e-6:
+                    for r in results:
+                        r['relevance_score'] = (r['relevance_score'] - min_score) / (max_score - min_score)
+
+            return results
+        except httpx.HTTPStatusError as e:
+            raise errors.RequesterError(f'Rerank request failed: {e.response.status_code} - {e.response.text}')
+        except httpx.TimeoutException:
+            raise errors.RequesterError('Rerank request timed out')
+        except Exception as e:
+            raise errors.RequesterError(f'Rerank request error: {str(e)}')
+
+    @staticmethod
+    def _parse_rerank_response(data: dict) -> typing.List[dict]:
+        """Parse rerank response from various providers.
+
+        Handles:
+        - Jina/Cohere/SiliconFlow: {"results": [{"index", "relevance_score"}]}
+        - Voyage AI: {"data": [{"index", "relevance_score"}]}
+        - DashScope: {"output": {"results": [{"index", "relevance_score"}]}}
+        """
+        if 'results' in data:
+            return data['results']
+        if 'data' in data:
+            return data['data']
+        if 'output' in data and isinstance(data['output'], dict):
+            return data['output'].get('results', [])
+        return []
@@ -25,6 +25,7 @@ spec:
  support_type:
  - llm
  - text-embedding
+  - rerank
  provider_category: manufacturer
 execution:
  python:
@@ -1,7 +1,8 @@
-<svg width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
-  <rect width="24" height="24" rx="5" fill="#7B68EE"/>
-  <circle cx="12" cy="12" r="6" fill="#FF6B35"/>
-  <circle cx="12" cy="12" r="3" fill="#7B68EE"/>
-  <path d="M12 6V18" stroke="#FFF" stroke-width="1.5" stroke-linecap="round"/>
-  <path d="M6 12H18" stroke="#FFF" stroke-width="1.5" stroke-linecap="round"/>
+<svg xmlns="http://www.w3.org/2000/svg" fill="none" viewBox="0 0 128 128" id="Chroma--Streamline-Svg-Logos" height="128" width="128">
+  <desc>
+    Chroma Streamline Icon: https://streamlinehq.com
+  </desc>
+  <path fill="#ffde2d" d="M84.88839999999999 104.10666666666665c23.0732 0 41.77773333333333 -17.956266666666664 41.77773333333333 -40.10653333333333 0 -22.150266666666667 -18.70453333333333 -40.10653333333333 -41.77773333333333 -40.10653333333333 -23.0732 0 -41.77773333333333 17.956266666666664 -41.77773333333333 40.10653333333333 0 22.150266666666667 18.70453333333333 40.10653333333333 41.77773333333333 40.10653333333333Z" stroke-width="1.3333"></path>
+  <path fill="#327eff" d="M43.111066666666666 104.10666666666665c23.0732 0 41.77773333333333 -17.956266666666664 41.77773333333333 -40.10653333333333 0 -22.150266666666667 -18.70453333333333 -40.10653333333333 -41.77773333333333 -40.10653333333333C20.037866666666666 23.8936 1.3333333333333333 41.849866666666664 1.3333333333333333 64.00013333333334 1.3333333333333333 86.15039999999999 20.037866666666666 104.10666666666665 43.111066666666666 104.10666666666665Z" stroke-width="1.3333"></path>
+  <path fill="#ff6446" d="M84.88866666666667 64.00013333333334c0 22.150399999999998 -18.704666666666665 40.10626666666666 -41.778 40.10626666666666V64.00013333333334h41.778Zm-41.778 0c0 -22.150266666666667 18.70453333333333 -40.10653333333333 41.778 -40.10653333333333v40.10653333333333H43.11066666666666Z" stroke-width="1.3333"></path>
 </svg>
@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Cohere</title><path clip-rule="evenodd" d="M8.128 14.099c.592 0 1.77-.033 3.398-.703 1.897-.781 5.672-2.2 8.395-3.656 1.905-1.018 2.74-2.366 2.74-4.18A4.56 4.56 0 0018.1 1H7.549A6.55 6.55 0 001 7.55c0 3.617 2.745 6.549 7.128 6.549z" fill="#39594D" fill-rule="evenodd"></path><path clip-rule="evenodd" d="M9.912 18.61a4.387 4.387 0 012.705-4.052l3.323-1.38c3.361-1.394 7.06 1.076 7.06 4.715a5.104 5.104 0 01-5.105 5.104l-3.597-.001a4.386 4.386 0 01-4.386-4.387z" fill="#D18EE2" fill-rule="evenodd"></path><path d="M4.776 14.962A3.775 3.775 0 001 18.738v.489a3.776 3.776 0 007.551 0v-.49a3.775 3.775 0 00-3.775-3.775z" fill="#FF7759"></path></svg>
@@ -0,0 +1,31 @@
+apiVersion: v1
+kind: LLMAPIRequester
+metadata:
+  name: cohere-rerank
+  label:
+    en_US: Cohere
+    zh_Hans: Cohere
+  icon: cohere.svg
+spec:
+  config:
+  - name: base_url
+    label:
+      en_US: Base URL
+      zh_Hans: 基础 URL
+    type: string
+    required: true
+    default: https://api.cohere.com/v2
+  - name: timeout
+    label:
+      en_US: Timeout
+      zh_Hans: 超时时间
+    type: integer
+    required: true
+    default: 120
+  support_type:
+  - rerank
+  provider_category: manufacturer
+execution:
+  python:
+    path: ./chatcmpl.py
+    attr: OpenAIChatCompletions
@@ -25,6 +25,7 @@ spec:
  support_type:
  - llm
  - text-embedding
+  - rerank
  provider_category: maas
 execution:
  python:
@@ -0,0 +1 @@
+<svg fill="currentColor" fill-rule="evenodd" height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Jina</title><path d="M6.608 21.416a4.608 4.608 0 100-9.217 4.608 4.608 0 000 9.217zM20.894 2.015c.614 0 1.106.492 1.106 1.106v9.002c0 5.13-4.148 9.309-9.217 9.37v-9.355l-.03-9.032c0-.614.491-1.106 1.106-1.106h7.158l-.123.015z"></path></svg>
@@ -0,0 +1,31 @@
+apiVersion: v1
+kind: LLMAPIRequester
+metadata:
+  name: jina-rerank
+  label:
+    en_US: Jina
+    zh_Hans: Jina
+  icon: jina.svg
+spec:
+  config:
+  - name: base_url
+    label:
+      en_US: Base URL
+      zh_Hans: 基础 URL
+    type: string
+    required: true
+    default: https://api.jina.ai/v1
+  - name: timeout
+    label:
+      en_US: Timeout
+      zh_Hans: 超时时间
+    type: integer
+    required: true
+    default: 120
+  support_type:
+  - rerank
+  provider_category: manufacturer
+execution:
+  python:
+    path: ./chatcmpl.py
+    attr: OpenAIChatCompletions
@@ -25,6 +25,7 @@ spec:
  support_type:
  - llm
  - text-embedding
+  - rerank
  provider_category: maas
 execution:
  python:
@@ -1,8 +1,17 @@
-<svg width="24" height="24" viewBox="0 0 24 24" fill="none" xmlns="http://www.w3.org/2000/svg">
-  <rect width="24" height="24" rx="5" fill="#1E3A5F"/>
-  <path d="M6 12C6 8.68629 8.68629 6 12 6C15.3137 6 18 8.68629 18 12" stroke="#4FC3F7" stroke-width="2" stroke-linecap="round"/>
-  <path d="M18 12C18 15.3137 15.3137 18 12 18C8.68629 18 6 15.3137 6 12" stroke="#81D4FA" stroke-width="2" stroke-linecap="round"/>
-  <circle cx="12" cy="12" r="2" fill="#4FC3F7"/>
-  <circle cx="6" cy="12" r="1.5" fill="#81D4FA"/>
-  <circle cx="18" cy="12" r="1.5" fill="#4FC3F7"/>
-</svg>
+<svg id="_图层_1" data-name="图层 1" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 334.84 76.22">
+  <defs>
+    <style>
+      .cls-1 {
+        fill: currentColor;
+      }
+    </style>
+  </defs>
+  <path class="cls-1" d="M308.56,23.63c-5.04,0-9.73,1.43-13.73,3.88V1.08l-12.56,4.61v70h12.56v-3.35c4,2.46,8.71,3.88,13.73,3.88,14.49,0,26.29-11.79,26.29-26.29s-11.79-26.29-26.29-26.29h0ZM308.56,63.88c-6.87,0-12.57-4.98-13.73-11.51v-4.91c1.16-6.54,6.88-11.51,13.73-11.51,7.7,0,13.96,6.26,13.96,13.96s-6.26,13.96-13.96,13.96Z"></path>
+  <path class="cls-1" d="M255.54,5.69v21.83c-4-2.46-8.71-3.88-13.73-3.88-14.49,0-26.29,11.79-26.29,26.29s11.79,26.29,26.29,26.29c5.04,0,9.73-1.43,13.73-3.88v3.35h12.56V1.08l-12.56,4.61ZM241.81,63.88c-7.7,0-13.96-6.26-13.96-13.96s6.26-13.96,13.96-13.96c6.87,0,12.57,4.98,13.73,11.51v4.91c-1.16,6.54-6.88,11.51-13.73,11.51Z"></path>
+  <polygon class="cls-1" points="195.35 52.2 186.65 61.17 200.64 75.62 209.32 75.62 218.01 75.62 195.35 52.2"></polygon>
+  <path class="cls-1" d="M167.14,4.59c.65,3.99.68,8.04.03,12.15-.03.17.16.3.31.21,3.82-2.21,7.82-3.69,12.01-4.33.12-.02.19-.13.17-.23-.68-4.13-.61-8.18-.03-12.16.02-.17-.16-.3-.31-.2-4.01,2.31-8.01,3.81-12.01,4.34-.12.01-.19.12-.17.23h0Z"></path>
+  <path class="cls-1" d="M198.75,24.09l-19.07,19.72v-25.57c-4.49.67-8.7,2.11-12.56,4.57v52.83h12.56v-13.87l3.78-3.9.02.02,8.68-8.97-.02-.02,23.98-24.8h-17.37Z"></path>
+  <path class="cls-1" d="M145.03,57.86c-2.56,4.45-7.17,7.2-12.13,7.2-5.96,0-11.3-3.96-13.32-9.85h38.87l.08-.42c.29-1.5.42-3.06.42-4.65,0-14.37-11.69-26.06-26.06-26.06s-26.06,11.69-26.06,26.06,11.69,26.06,26.06,26.06c9.63,0,18.43-5.28,22.98-13.77l.26-.49-11.1-4.08h-.01ZM132.88,35.19h.03c5.96,0,11.3,3.96,13.32,9.85h-26.67c2.02-5.89,7.36-9.85,13.32-9.85Z"></path>
+  <path class="cls-1" d="M75.92,65.07c-5.96,0-11.29-3.96-13.32-9.85h38.87l.08-.42c.29-1.5.42-3.06.42-4.65,0-14.37-11.69-26.06-26.06-26.06s-26.06,11.69-26.06,26.06,11.69,26.06,26.06,26.06c9.63,0,18.43-5.28,22.98-13.77l.26-.49h0l-11.1-4.08c-2.56,4.45-7.17,7.2-12.13,7.2h-.01ZM75.92,35.19h.03c5.96,0,11.29,3.96,13.32,9.85h-26.67c2.03-5.89,7.36-9.85,13.32-9.85Z"></path>
+  <path class="cls-1" d="M30.43,45.58l-10.2-1.91c-3.03-.56-4.98-2.25-4.98-4.33,0-1.5,1.61-4.35,7.68-4.35,5.53,0,9.36,3.5,10.25,6.26l10.9-4-.14-.42c-1.17-3.54-3.5-6.58-6.94-9.04-3.49-2.49-8.04-3.69-13.88-3.69s-10.98,1.5-14.78,4.34c-3.88,2.91-5.84,6.76-5.84,11.46,0,7.98,4.72,12.77,14.42,14.64l9.9,1.81c3.05.61,4.94,2.27,4.94,4.33,0,2.61-3.58,4.44-8.7,4.44-5.79,0-9.9-3.72-11.85-7.14L0,62.1l.14.39c1.3,3.8,3.89,7.07,7.7,9.71,3.78,2.6,8.65,3.95,14.51,3.98l.25.03c6.87,0,12.55-1.57,16.43-4.53,3.98-3.05,6-6.99,6-11.74,0-3.73-1.14-6.7-3.6-9.33-2.27-2.42-5.98-4.11-10.98-5.02h-.02Z"></path>
+</svg>
@@ -25,6 +25,7 @@ spec:
  support_type:
  - llm
  - text-embedding
+  - rerank
  provider_category: maas
 execution:
  python:
@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Voyage</title><path d="M5.407 0v.066a.974.974 0 00-.048.245c-.011.11-.016.208-.016.295 0 .339.043.715.128 1.13.097.405.274.912.531 1.524l7.125 16.366L20.011 3.39c.161-.404.333-.846.515-1.327.182-.48.273-.966.273-1.458a1.406 1.406 0 00-.096-.54V0H24v.066c-.204.207-.45.578-.74 1.114-.29.535-.606 1.195-.949 1.982L13.095 24h-1.287L3.075 3.965c-.204-.47-.418-.923-.644-1.36-.214-.437-.418-.83-.61-1.18-.194-.36-.365-.66-.515-.9A5.666 5.666 0 001 .064V0h4.407z" fill="#012E33"></path></svg>
@@ -0,0 +1,31 @@
+apiVersion: v1
+kind: LLMAPIRequester
+metadata:
+  name: voyageai-rerank
+  label:
+    en_US: Voyage AI
+    zh_Hans: Voyage AI
+  icon: voyageai.svg
+spec:
+  config:
+  - name: base_url
+    label:
+      en_US: Base URL
+      zh_Hans: 基础 URL
+    type: string
+    required: true
+    default: https://api.voyageai.com/v1
+  - name: timeout
+    label:
+      en_US: Timeout
+      zh_Hans: 超时时间
+    type: integer
+    required: true
+    default: 120
+  support_type:
+  - rerank
+  provider_category: manufacturer
+execution:
+  python:
+    path: ./chatcmpl.py
+    attr: OpenAIChatCompletions
@@ -172,6 +172,45 @@ class LocalAgentRunner(runner.RequestRunner):
                if result:
                    all_results.extend(result)

+            # Rerank step: re-score results using a rerank model if configured
+            local_agent_config = query.pipeline_config.get('ai', {}).get('local-agent', {})
+            rerank_model_uuid = local_agent_config.get('rerank-model', '')
+            if rerank_model_uuid == '__none__':
+                rerank_model_uuid = ''
+            self.ap.logger.info(
+                f'Rerank config: model_uuid={rerank_model_uuid!r}, '
+                f'results={len(all_results)}, '
+                f'local_agent_keys={list(local_agent_config.keys())}'
+            )
+            if all_results and rerank_model_uuid:
+                try:
+                    rerank_model = await self.ap.model_mgr.get_rerank_model_by_uuid(rerank_model_uuid)
+                    rerank_top_k = int(local_agent_config.get('rerank-top-k', 5))
+
+                    doc_texts = []
+                    for entry in all_results:
+                        text = ' '.join(c.text for c in entry.content if c.type == 'text' and c.text)
+                        doc_texts.append(text)
+
+                    doc_texts_capped = doc_texts[:64]
+                    scores = await rerank_model.provider.invoke_rerank(
+                        model=rerank_model,
+                        query=user_message_text,
+                        documents=doc_texts_capped,
+                    )
+
+                    scored = sorted(scores, key=lambda x: x.get('relevance_score', 0), reverse=True)
+                    top_indices = [s['index'] for s in scored[:rerank_top_k] if s['index'] < len(all_results)]
+                    all_results = [all_results[i] for i in top_indices]
+
+                    self.ap.logger.info(
+                        f'Rerank complete: {len(doc_texts)} docs reranked -> top {len(all_results)} kept (top_k={rerank_top_k})'
+                    )
+                except ValueError:
+                    self.ap.logger.warning(f'Rerank model {rerank_model_uuid} not found, skipping rerank')
+                except Exception as e:
+                    self.ap.logger.warning(f'Rerank failed, using original order: {e}')
+
            final_user_message_text = ''

            if all_results:
				`@@ -0,0 +1 @@`
				<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Cohere</title><path clip-rule="evenodd" d="M8.128 14.099c.592 0 1.77-.033 3.398-.703 1.897-.781 5.672-2.2 8.395-3.656 1.905-1.018 2.74-2.366 2.74-4.18A4.56 4.56 0 0018.1 1H7.549A6.55 6.55 0 001 7.55c0 3.617 2.745 6.549 7.128 6.549z" fill="#39594D" fill-rule="evenodd"></path><path clip-rule="evenodd" d="M9.912 18.61a4.387 4.387 0 012.705-4.052l3.323-1.38c3.361-1.394 7.06 1.076 7.06 4.715a5.104 5.104 0 01-5.105 5.104l-3.597-.001a4.386 4.386 0 01-4.386-4.387z" fill="#D18EE2" fill-rule="evenodd"></path><path d="M4.776 14.962A3.775 3.775 0 001 18.738v.489a3.776 3.776 0 007.551 0v-.49a3.775 3.775 0 00-3.775-3.775z" fill="#FF7759"></path></svg>
				`@@ -0,0 +1 @@`
				`<svg fill="currentColor" fill-rule="evenodd" height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Jina</title><path d="M6.608 21.416a4.608 4.608 0 100-9.217 4.608 4.608 0 000 9.217zM20.894 2.015c.614 0 1.106.492 1.106 1.106v9.002c0 5.13-4.148 9.309-9.217 9.37v-9.355l-.03-9.032c0-.614.491-1.106 1.106-1.106h7.158l-.123.015z"></path></svg>`