Skip to content

CUDA: faster q2_K, q3_K MMQ + int8 tensor cores#7921

Merged
JohannesGaessler merged 6 commits intoggerganov:masterfrom JohannesGaessler:cuda-ptx-mma-17Jun 14, 2024

Commits

Commits on Jun 13, 2024

Commits on Jun 14, 2024