[v1] EngineArgs for better config handling for v1 (#10382)

Signed-off-by: rickyx <rickyx@anyscale.com>
2024-11-25 21:09:43 -08:00
parent a6760f6456
commit 519e8e4182
13 changed files with 109 additions and 27 deletions
--- a/vllm/entrypoints/openai/api_server.py
+++ b/vllm/entrypoints/openai/api_server.py
@@ -135,8 +135,8 @@ async def build_async_engine_client_from_engine_args(
    # TODO: fill out feature matrix.
    if (MQLLMEngineClient.is_unsupported_config(engine_args)
            or envs.VLLM_USE_V1 or disable_frontend_multiprocessing):
-
-        engine_config = engine_args.create_engine_config()
+        engine_config = engine_args.create_engine_config(
+            UsageContext.OPENAI_API_SERVER)
        uses_ray = getattr(AsyncLLMEngine._get_executor_cls(engine_config),
                           "uses_ray", False)