[Misc] Update request-extras parity for batch chat completion (#47333)
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com>
Taneem Ibrahim committed
Jul 4, 2026 at 09:19 UTC
2f21224527bdac2e4ccd251563101257ad61b214
2 files changed
+5
-1
vllm/entrypoints/openai/chat_completion/batch_serving.py
+1
-1
@@ -181,7 +181,7 @@ class OpenAIServingChatBatch(OpenAIServingChat):
181
sub_request_id,
182
lora_request=lora_request,
183
trace_headers=trace_headers,
184
- priority=request.priority if hasattr(request, "priority") else 0,
184
+ priority=request.priority,
185
data_parallel_rank=data_parallel_rank,
186
reasoning_ended=None,
187
)
vllm/entrypoints/openai/chat_completion/protocol.py
+4
@@ -1020,6 +1020,10 @@ class BatchChatCompletionRequest(OpenAIBaseModel):
1020
continue_final_message: bool = False
1021
chat_template: str | None = None
1022
chat_template_kwargs: dict[str, Any] | None = None
1023
+ media_io_kwargs: dict[str, dict[str, Any]] | None = None
1024
+ mm_processor_kwargs: dict[str, Any] | None = None
1025
+ priority: int = Field(default=0, ge=_INT64_MIN, le=_INT64_MAX)
1026
+ cache_salt: str | None = None
1027
include_stop_str_in_output: bool = False
1028
guided_decoding_backend: str | None = None
1029
echo: bool = False