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 16, 04:30 UTC
[Doc] Fix typo in the help message of '--guided-decoding-backend' (#11440)
yansh97 committed
Dec 24, 2024 at 04:20 UTC
94d545a1a18e20ea8763a6760194589b8a3c9065
1 file changed
+1
-1
vllm/engine/arg_utils.py
+1
-1
@@ -373,7 +373,7 @@ class EngineArgs:
373
choices=['outlines', 'lm-format-enforcer', 'xgrammar'],
374
help='Which engine will be used for guided decoding'
375
' (JSON schema / regex etc) by default. Currently support '
376
- 'https://github.com/outlines-dev/outlines,'
376
+ 'https://github.com/outlines-dev/outlines, '
377
'https://github.com/mlc-ai/xgrammar, and '
378
'https://github.com/noamgat/lm-format-enforcer.'
379
' Can be overridden per request via guided_decoding_backend'