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 6, 17:00 UTC
[Model Runner V2] Fix processed logits in sample() (#37144)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Woosuk Kwon committed
Mar 16, 2026 at 00:35 UTC
96efb91480cd973dbcffab25ccb4b3a119b9929e
1 file changed
+1
-1
vllm/v1/worker/gpu/sample/sampler.py
+1
-1
@@ -178,4 +178,4 @@ class Sampler:
178
pos,
179
apply_temperature=False,
180
)
181
- return sampled, logits
181
+ return sampled, processed_logits