vllm/v1/worker/gpu_input_batch.py · 2f385183f35497e030ef22c9820d83b83bc4f6db · OpenDAS / vllm_cscc · GitLab

Find file Blame History Permalink

[V1][Minor] Set pin_memory=False for token_ids_cpu tensor (#11581) · 42bb201f
Woosuk Kwon authored Dec 28, 2024
```
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
```
42bb201f

gpu_input_batch.py 17.8 KB