From 4ebf61ebda8dc8bf090bd4ee14d2855f694111b0 Mon Sep 17 00:00:00 2001 From: Taneem Ibrahim Date: Wed, 9 Sep 2026 17:57:01 -0500 Subject: [PATCH] [Cohere] Bound remaining request priorities to the MessagePack int64 range (#56146) Signed-off-by: Taneem Ibrahim --- vllm/entrypoints/cohere/protocol.py | 2 +- vllm/entrypoints/generate/generative_scoring/serving.py | 2 ++ vllm/entrypoints/pooling/embed/protocol.py | 2 +- 3 files changed, 4 insertions(+), 2 deletions(-) diff --git a/vllm/entrypoints/cohere/protocol.py b/vllm/entrypoints/cohere/protocol.py index 483216a699..1432e6eb11 100644 --- a/vllm/entrypoints/cohere/protocol.py +++ b/vllm/entrypoints/cohere/protocol.py @@ -170,7 +170,7 @@ class CohereChatV2Request(BaseModel): thinking: Thinking | None = None # Scheduling - priority: int | None = None + priority: int | None = Field(default=None, ge=-(2**63), le=2**63 - 1) # vLLM-specific extensions (not in Cohere spec). These mirror what the # Anthropic and OpenAI surfaces already expose so V2 callers can reach diff --git a/vllm/entrypoints/generate/generative_scoring/serving.py b/vllm/entrypoints/generate/generative_scoring/serving.py index 59db919225..753054d1b8 100644 --- a/vllm/entrypoints/generate/generative_scoring/serving.py +++ b/vllm/entrypoints/generate/generative_scoring/serving.py @@ -93,6 +93,8 @@ class GenerativeScoringRequest(OpenAIBaseModel): ) priority: int = Field( default=0, + ge=-(2**63), + le=2**63 - 1, description=( "The priority of the request (lower means earlier handling; default: 0)." ), diff --git a/vllm/entrypoints/pooling/embed/protocol.py b/vllm/entrypoints/pooling/embed/protocol.py index c9a423d8a7..a853a8ea00 100644 --- a/vllm/entrypoints/pooling/embed/protocol.py +++ b/vllm/entrypoints/pooling/embed/protocol.py @@ -236,7 +236,7 @@ class CohereEmbedRequest(BaseModel): embedding_types: list[CohereEmbeddingType] | None = None truncate: CohereTruncate = "END" max_tokens: int | None = None - priority: int = 0 + priority: int = Field(default=0, ge=-(2**63), le=2**63 - 1) @model_validator(mode="after") def validate_input_fields(self):