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, 19:30 UTC
[Doc] Super tiny fix doc typo (#6949)
fzyzcjy committed
Jul 31, 2024 at 00:14 UTC
f05840368335aa9c8184239d3c8bd986e44692f7
1 file changed
+3
-3
vllm/engine/arg_utils.py
+3
-3
@@ -632,9 +632,9 @@ class EngineArgs:
632
'--preemption-mode',
633
type=str,
634
default=None,
635
- help='If \'recompute\', the engine performs preemption by block '
636
- 'swapping; If \'swap\', the engine performs preemption by block '
637
- 'swapping.')
635
+ help='If \'recompute\', the engine performs preemption by '
636
+ 'recomputing; If \'swap\', the engine performs preemption by '
637
+ 'block swapping.')
638
639
parser.add_argument(
640
"--served-model-name",