A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 09:45 UTC
[Misc] Update deprecation message for `--enable-reasoning` (#18404)
Hyogeun Oh (오효근) committed
May 21, 2025 at 23:33 UTC
2b161045574fc4ba23052c6578ad8f97090db884
1 file changed
+1
-1
vllm/engine/arg_utils.py
+1
-1
@@ -577,7 +577,7 @@ class EngineArgs:
577
action=argparse.BooleanOptionalAction,
578
deprecated=True,
579
help="[DEPRECATED] The `--enable-reasoning` flag is deprecated as "
580
- "of v0.8.6. Use `--reasoning-parser` to specify the reasoning "
580
+ "of v0.9.0. Use `--reasoning-parser` to specify the reasoning "
581
"parser backend instead. This flag (`--enable-reasoning`) will be "
582
"removed in v0.10.0. When `--reasoning-parser` is specified, "
583
"reasoning mode is automatically enabled.")