chore: indicate to mypy that InferenceProvider.rerank is concrete (#3238)

Author: mattf

llama_stack/apis/inference/inference.py

@@ -1170,6 +1170,7 @@ async def rerank(
         :returns: RerankResponse with indices sorted by relevance score (descending).
         """
         raise NotImplementedError("Reranking is not implemented")
+        return  # this is so mypy's safe-super rule will consider the method concrete
 
     @webmethod(route="/openai/v1/completions", method="POST")
     async def openai_completion(

llama_stack/providers/inline/inference/meta_reference/inference.py

@@ -33,9 +33,6 @@
     InterleavedContent,
     LogProbConfig,
     Message,
-    OpenAIChatCompletionContentPartImageParam,
-    OpenAIChatCompletionContentPartTextParam,
-    RerankResponse,
     ResponseFormat,
     SamplingParams,
     StopReason,
@@ -445,15 +442,6 @@ async def batch_chat_completion(
         results = await self._nonstream_chat_completion(request_batch)
         return BatchChatCompletionResponse(batch=results)
 
-    async def rerank(
-        self,
-        model: str,
-        query: str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam,
-        items: list[str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam],
-        max_num_results: int | None = None,
-    ) -> RerankResponse:
-        raise NotImplementedError("Reranking is not supported for Meta Reference")
-
     async def _nonstream_chat_completion(
         self, request_batch: list[ChatCompletionRequest]
     ) -> list[ChatCompletionResponse]:

llama_stack/providers/inline/inference/sentence_transformers/sentence_transformers.py

@@ -12,9 +12,6 @@
     InterleavedContent,
     LogProbConfig,
     Message,
-    OpenAIChatCompletionContentPartImageParam,
-    OpenAIChatCompletionContentPartTextParam,
-    RerankResponse,
     ResponseFormat,
     SamplingParams,
     ToolChoice,
@@ -125,12 +122,3 @@ async def batch_chat_completion(
         logprobs: LogProbConfig | None = None,
     ):
         raise NotImplementedError("Batch chat completion is not supported for Sentence Transformers")
-
-    async def rerank(
-        self,
-        model: str,
-        query: str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam,
-        items: list[str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam],
-        max_num_results: int | None = None,
-    ) -> RerankResponse:
-        raise NotImplementedError("Reranking is not supported for Sentence Transformers")

llama_stack/providers/remote/inference/llama_openai_compat/llama.py

@@ -3,11 +3,6 @@
 #
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
-from llama_stack.apis.inference import (
-    OpenAIChatCompletionContentPartImageParam,
-    OpenAIChatCompletionContentPartTextParam,
-    RerankResponse,
-)
 from llama_stack.log import get_logger
 from llama_stack.providers.remote.inference.llama_openai_compat.config import LlamaCompatConfig
 from llama_stack.providers.utils.inference.litellm_openai_mixin import LiteLLMOpenAIMixin
@@ -59,12 +54,3 @@ async def initialize(self):
 
     async def shutdown(self):
         await super().shutdown()
-
-    async def rerank(
-        self,
-        model: str,
-        query: str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam,
-        items: list[str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam],
-        max_num_results: int | None = None,
-    ) -> RerankResponse:
-        raise NotImplementedError("Reranking is not supported for Llama OpenAI Compat")

llama_stack/providers/remote/inference/ollama/ollama.py

@@ -37,14 +37,11 @@
     Message,
     OpenAIChatCompletion,
     OpenAIChatCompletionChunk,
-    OpenAIChatCompletionContentPartImageParam,
-    OpenAIChatCompletionContentPartTextParam,
     OpenAICompletion,
     OpenAIEmbeddingsResponse,
     OpenAIEmbeddingUsage,
     OpenAIMessageParam,
     OpenAIResponseFormatParam,
-    RerankResponse,
     ResponseFormat,
     SamplingParams,
     TextTruncation,
@@ -644,15 +641,6 @@ async def batch_chat_completion(
     ):
         raise NotImplementedError("Batch chat completion is not supported for Ollama")
 
-    async def rerank(
-        self,
-        model: str,
-        query: str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam,
-        items: list[str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam],
-        max_num_results: int | None = None,
-    ) -> RerankResponse:
-        raise NotImplementedError("Reranking is not supported for Ollama")
-
 
 async def convert_message_to_openai_dict_for_ollama(message: Message) -> list[dict]:
     async def _convert_content(content) -> dict:

llama_stack/providers/remote/inference/vllm/vllm.py

@@ -39,15 +39,12 @@
     Message,
     ModelStore,
     OpenAIChatCompletion,
-    OpenAIChatCompletionContentPartImageParam,
-    OpenAIChatCompletionContentPartTextParam,
     OpenAICompletion,
     OpenAIEmbeddingData,
     OpenAIEmbeddingsResponse,
     OpenAIEmbeddingUsage,
     OpenAIMessageParam,
     OpenAIResponseFormatParam,
-    RerankResponse,
     ResponseFormat,
     SamplingParams,
     TextTruncation,
@@ -736,12 +733,3 @@ async def batch_chat_completion(
         logprobs: LogProbConfig | None = None,
     ):
         raise NotImplementedError("Batch chat completion is not supported for vLLM")
-
-    async def rerank(
-        self,
-        model: str,
-        query: str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam,
-        items: list[str | OpenAIChatCompletionContentPartTextParam | OpenAIChatCompletionContentPartImageParam],
-        max_num_results: int | None = None,
-    ) -> RerankResponse:
-        raise NotImplementedError("Reranking is not supported for vLLM")