[Core] [Frontend] Priority scheduling for embeddings and in the OpenAI-API (#8965)

vllm/engine/async_llm_engine.py

@@ -1043,6 +1043,7 @@ class AsyncLLMEngine:
         request_id: str,
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
+        priority: int = 0,
     ) -> AsyncGenerator[EmbeddingRequestOutput, None]:
         """Generate outputs for a request from an embedding model.
 
@@ -1057,6 +1058,8 @@ class AsyncLLMEngine:
             request_id: The unique id of the request.
             lora_request: LoRA request to use for generation, if any.
             trace_headers: OpenTelemetry trace headers.
+            priority: The priority of the request.
+                Only applicable with priority scheduling.
 
         Yields:
             The output `EmbeddingRequestOutput` objects from the LLMEngine
@@ -1109,6 +1112,7 @@ class AsyncLLMEngine:
                 pooling_params,
                 lora_request=lora_request,
                 trace_headers=trace_headers,
+                priority=priority,
         ):
             yield LLMEngine.validate_output(output, EmbeddingRequestOutput)
 

vllm/engine/multiprocessing/__init__.py

@@ -30,6 +30,7 @@ class RPCProcessRequest:
     lora_request: Optional[LoRARequest] = None
     trace_headers: Optional[Mapping[str, str]] = None
     prompt_adapter_request: Optional[PromptAdapterRequest] = None
+    priority: int = 0
 
     @overload  # DEPRECATED
     def __init__(
@@ -41,6 +42,7 @@ class RPCProcessRequest:
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
         prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+        priority: int = 0,
     ) -> None:
         ...
 
@@ -53,6 +55,7 @@ class RPCProcessRequest:
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
         prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+        priority: int = 0,
     ) -> None:
         ...
 
@@ -68,6 +71,7 @@ class RPCProcessRequest:
             lora_request: Optional[LoRARequest] = None,
             trace_headers: Optional[Mapping[str, str]] = None,
             prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+            priority: int = 0,
             *,
             inputs: Optional[PromptType] = None,  # DEPRECATED
     ) -> None:
@@ -84,6 +88,7 @@ class RPCProcessRequest:
         self.lora_request = lora_request
         self.trace_headers = trace_headers
         self.prompt_adapter_request = prompt_adapter_request
+        self.priority = priority
 
 
 @dataclass

vllm/engine/multiprocessing/client.py

@@ -380,6 +380,7 @@ class MQLLMEngineClient:
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
         prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+        priority: int = 0,
     ) -> AsyncGenerator[RequestOutput, None]:
         ...
 
@@ -392,6 +393,7 @@ class MQLLMEngineClient:
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
         prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+        priority: int = 0,
     ) -> AsyncGenerator[RequestOutput, None]:
         ...
 
@@ -407,6 +409,7 @@ class MQLLMEngineClient:
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
         prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+        priority: int = 0,
         *,
         inputs: Optional[PromptType] = None  # DEPRECATED
     ) -> AsyncGenerator[RequestOutput, None]:
@@ -425,6 +428,9 @@ class MQLLMEngineClient:
             trace_headers: OpenTelemetry trace headers.
             prompt_adapter_request: Prompt Adapter request to use
                                             for generation, if any.
+            priority: Priority of the request (lower means earlier handling). 
+                Any priority other than 0 will lead to an error if the 
+                scheduling policy is not "priority".
         """
         if inputs is not None:
             prompt = inputs
@@ -433,7 +439,7 @@ class MQLLMEngineClient:
 
         return self._process_request(prompt, sampling_params, request_id,
                                      lora_request, trace_headers,
-                                     prompt_adapter_request)
+                                     prompt_adapter_request, priority)
 
     @overload  # DEPRECATED
     def encode(
@@ -444,6 +450,7 @@ class MQLLMEngineClient:
         request_id: str,
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
+        priority: int = 0,
     ) -> AsyncGenerator[EmbeddingRequestOutput, None]:
         ...
 
@@ -455,6 +462,7 @@ class MQLLMEngineClient:
         request_id: str,
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
+        priority: int = 0,
     ) -> AsyncGenerator[EmbeddingRequestOutput, None]:
         ...
 
@@ -469,6 +477,7 @@ class MQLLMEngineClient:
         request_id: Optional[str] = None,
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
+        priority: int = 0,
         *,
         inputs: Optional[PromptType] = None  # DEPRECATED
     ) -> AsyncGenerator[EmbeddingRequestOutput, None]:
@@ -496,7 +505,7 @@ class MQLLMEngineClient:
                 and request_id is not None)
 
         return self._process_request(prompt, pooling_params, request_id,
-                                     lora_request, trace_headers)
+                                     lora_request, trace_headers, priority)
 
     async def _process_request(
         self,
@@ -505,7 +514,8 @@ class MQLLMEngineClient:
         request_id: str,
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
-        prompt_adapter_request: Optional[PromptAdapterRequest] = None
+        prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+        priority: int = 0,
     ) -> Union[AsyncGenerator[RequestOutput, None], AsyncGenerator[
             EmbeddingRequestOutput, None]]:
         """Send an RPCGenerateRequest to the RPCServer and stream responses."""
@@ -550,7 +560,9 @@ class MQLLMEngineClient:
                     request_id=request_id,
                     lora_request=lora_request,
                     trace_headers=trace_headers,
-                    prompt_adapter_request=prompt_adapter_request))
+                    prompt_adapter_request=prompt_adapter_request,
+                    priority=priority,
+                ))
 
             # 3) Send the RPCGenerateRequest to the MQLLMEngine.
             parts = (request_bytes,

vllm/engine/protocol.py

@@ -40,7 +40,8 @@ class EngineClient(Protocol):
         request_id: str,
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
-        prompt_adapter_request: Optional[PromptAdapterRequest] = None
+        prompt_adapter_request: Optional[PromptAdapterRequest] = None,
+        priority: int = 0,
     ) -> AsyncGenerator[RequestOutput, None]:
         """Generate outputs for a request."""
         ...
@@ -52,6 +53,7 @@ class EngineClient(Protocol):
         request_id: str,
         lora_request: Optional[LoRARequest] = None,
         trace_headers: Optional[Mapping[str, str]] = None,
+        priority: int = 0,
     ) -> AsyncGenerator[EmbeddingRequestOutput, None]:
         """Generate outputs for a request from an embedding model."""
         ...

vllm/entrypoints/openai/protocol.py

@@ -279,6 +279,12 @@ class ChatCompletionRequest(OpenAIBaseModel):
         description=(
             "If specified, will override the default whitespace pattern "
             "for guided json decoding."))
+    priority: int = Field(
+        default=0,
+        description=(
+            "The priority of the request (lower means earlier handling; "
+            "default: 0). Any priority other than 0 will raise an error "
+            "if the served model does not use priority scheduling."))
 
     # doc: end-chat-completion-extra-params
 
@@ -552,6 +558,12 @@ class CompletionRequest(OpenAIBaseModel):
         description=(
             "If specified, will override the default whitespace pattern "
             "for guided json decoding."))
+    priority: int = Field(
+        default=0,
+        description=(
+            "The priority of the request (lower means earlier handling; "
+            "default: 0). Any priority other than 0 will raise an error "
+            "if the served model does not use priority scheduling."))
 
     # doc: end-completion-extra-params
 
@@ -665,6 +677,16 @@ class EmbeddingRequest(OpenAIBaseModel):
 
     # doc: end-embedding-pooling-params
 
+    # doc: begin-embedding-extra-params
+    priority: int = Field(
+        default=0,
+        description=(
+            "The priority of the request (lower means earlier handling; "
+            "default: 0). Any priority other than 0 will raise an error "
+            "if the served model does not use priority scheduling."))
+
+    # doc: end-embedding-extra-params
+
     def to_pooling_params(self):
         return PoolingParams(additional_data=self.additional_data)
 

vllm/entrypoints/openai/serving_chat.py

@@ -235,6 +235,7 @@ class OpenAIServingChat(OpenAIServing):
                 lora_request=lora_request,
                 trace_headers=trace_headers,
                 prompt_adapter_request=prompt_adapter_request,
+                priority=request.priority,
             )
         except ValueError as e:
             # TODO: Use a vllm-specific Validation Error

vllm/entrypoints/openai/serving_completion.py

@@ -148,6 +148,7 @@ class OpenAIServingCompletion(OpenAIServing):
                     lora_request=lora_request,
                     prompt_adapter_request=prompt_adapter_request,
                     trace_headers=trace_headers,
+                    priority=request.priority,
                 )
 
                 generators.append(generator)

vllm/entrypoints/openai/serving_embedding.py

@@ -148,6 +148,7 @@ class OpenAIServingEmbedding(OpenAIServing):
                     pooling_params,
                     request_id_item,
                     lora_request=lora_request,
+                    priority=request.priority,
                 )
 
                 generators.append(generator)