[Bugfix] Reject negative values for max_logprobs and long_prefill_token_threshold (#44002)
Signed-off-by: jwzheng96 <jianweizheng@pku.edu.cn> Signed-off-by: JianweiZheng <32029023+jwzheng96@users.noreply.github.com> Co-authored-by: Claude <noreply@anthropic.com> Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com> Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
JianweiZheng committed
Jun 30, 2026 at 20:01 UTC
1907d3854ae7dd0e1a5755fafbe93165ae6413a6
1 file changed
+1
-1
vllm/config/scheduler.py
+1
-1
@@ -79,7 +79,7 @@ class SchedulerConfig:
79
80
long_prefill_token_threshold: int = Field(default=0, ge=0)
81
"""For chunked prefill, a request is considered long if the prompt is
82
- longer than this number of tokens."""
82
+ longer than this number of tokens. 0 disables the cap (default)."""
83
84
enable_chunked_prefill: bool = True
85
"""If True, prefill requests can be chunked based