vllm/v1/worker/gpu_model_runner.py · ed2f282bc8cacfbe49a86ab1dbaf7e4471ec6b18 · OpenDAS / vllm_cscc · GitLab

Find file Blame History Permalink

[Perf] Optimize redundant sync for pooling model, 3.7% Throughput Improvement (#39113) · ed2f282b
Wentao Ye authored Apr 09, 2026
```
Signed-off-by: yewentao256 <zhyanwentao@126.com>
```
ed2f282b

gpu_model_runner.py 304 KB