Skip to content

[ROCm] Optimize kgemm_4bit_inference_naive for ROCm, use it for batch sizes other than 1 #255

[ROCm] Optimize kgemm_4bit_inference_naive for ROCm, use it for batch sizes other than 1

[ROCm] Optimize kgemm_4bit_inference_naive for ROCm, use it for batch sizes other than 1 #255

Annotations

1 error and 1 warning

CUDA (windows, T4, 11.8.0, 2.7.1, https://download.pytorch.org/whl/cu118)  /  test

failed Apr 22, 2026 in 1m 46s