vllm/v1/worker/gpu_input_batch.py · 995dea1354cbbb378f837b6897af420fd2f75322 · OpenDAS / vllm_cscc

[Perf] Remove redundant device copies for CPU-only pooling token IDs, 48.9%... · 995dea13

Wentao Ye authored Mar 29, 2026


[Perf] Remove redundant device copies for CPU-only pooling token IDs, 48.9% E2E throughput improvement (#38139)
Signed-off-by: yewentao256 <zhyanwentao@126.com>

995dea13

gpu_input_batch.py 43.8 KB

Replace gpu_input_batch.py