vllm/v1/worker/tpu_model_runner.py · 2a84fb422fc62ab29238dccbf7bdb214fc51c31e · OpenDAS / vllm_cscc

[TPU] kv cache update kernel doesn't need to be padded slices to multiple of... · 2a84fb42

Chengji Yao authored Aug 09, 2025


[TPU] kv cache update kernel doesn't need to be padded slices to multiple of num_slices_per_block (#22394)
Signed-off-by: Chengji Yao <chengjiyao@gmail.com>
Co-authored-by: Chengji Yao <chengjiyao@gmail.com>

2a84fb42

tpu_model_runner.py 91.9 KB

Replace tpu_model_runner.py