xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-06-01 04:17:03 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Fix wrong CLI defaults for dynamic SchedulerConfig fields (#28872)

Browse Source 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung 2025-11-18 12:30:29 +08:00 committed by GitHub
parent 3ddcf46011
commit bf9e1e8767
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 17 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
tests/entrypoints/openai/test_enable_force_include_usage.py
View File

@ -17,7 +17,7 @@ def chat_server_with_force_include_usage(request): # noqa: F811
"128", "128",
"--enforce-eager", "--enforce-eager",
"--max-num-seqs", "--max-num-seqs",
"1", "4",
"--enable-force-include-usage", "--enable-force-include-usage",
"--port", "--port",
"55857", "55857",
@ -78,7 +78,7 @@ def transcription_server_with_force_include_usage():
"--dtype", "--dtype",
"bfloat16", "bfloat16",
"--max-num-seqs", "--max-num-seqs",
"1", "4",
"--enforce-eager", "--enforce-eager",
"--enable-force-include-usage", "--enable-force-include-usage",
"--gpu-memory-utilization", "--gpu-memory-utilization",

18
vllm/engine/arg_utils.py
View File

@ -1046,10 +1046,18 @@ class EngineArgs:
description=SchedulerConfig.__doc__, description=SchedulerConfig.__doc__,
) )
scheduler_group.add_argument( scheduler_group.add_argument(
"--max-num-batched-tokens", **scheduler_kwargs["max_num_batched_tokens"] "--max-num-batched-tokens",
**{
**scheduler_kwargs["max_num_batched_tokens"],
"default": None,
},
) )
scheduler_group.add_argument( scheduler_group.add_argument(
"--max-num-seqs", **scheduler_kwargs["max_num_seqs"] "--max-num-seqs",
**{
**scheduler_kwargs["max_num_seqs"],
"default": None,
},
) )
scheduler_group.add_argument( scheduler_group.add_argument(
"--max-num-partial-prefills", **scheduler_kwargs["max_num_partial_prefills"] "--max-num-partial-prefills", **scheduler_kwargs["max_num_partial_prefills"]
@ -1071,7 +1079,11 @@ class EngineArgs:
"--scheduling-policy", **scheduler_kwargs["policy"] "--scheduling-policy", **scheduler_kwargs["policy"]
) )
scheduler_group.add_argument( scheduler_group.add_argument(
"--enable-chunked-prefill", **scheduler_kwargs["enable_chunked_prefill"] "--enable-chunked-prefill",
**{
**scheduler_kwargs["enable_chunked_prefill"],
"default": None,
},
) )
scheduler_group.add_argument( scheduler_group.add_argument(
"--disable-chunked-mm-input", **scheduler_kwargs["disable_chunked_mm_input"] "--disable-chunked-mm-input", **scheduler_kwargs["disable_chunked_mm_input"]