4 Commits

Author	SHA1	Message	Date
Forkoz	58998acc9f	Fix cuda kernel for Pascal & Cuda 6/6.1 ... When I left the other functions to use normal atomic add it seemed like a small speedup. 4.79 it/s vs 5.23 it/s	2023-03-23 07:33:57 -05:00
John Smith	3471be4e56	add fast_4bit_matmul and auto switch 2 methods according to bottleneck	2023-03-21 08:43:07 +00:00
John Smith	5b64833390	add half support on cuda kernel	2023-03-20 09:19:05 +00:00
John Smith	551f62a0e8	add patch for gptq and peft	2023-03-18 13:31:48 +08:00