Philipp Moritz 12628d3c78 [Kernel] Optimize FP8 support for MoE kernel / Mixtral via static scales (#4343) ...

Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>

2024-04-27 04:49:59 +00:00

..

aqlm

AQLM CUDA support (#3287)

2024-04-23 13:59:33 -04:00

awq

Refactor 2 awq gemm kernels into m16nXk32 (#2723)

2024-02-12 11:02:17 -08:00

fp8

[Kernel] Optimize FP8 support for MoE kernel / Mixtral via static scales (#4343)

2024-04-27 04:49:59 +00:00

fp8_e5m2_kvcache

Fix compile error when using rocm (#2648)

2024-02-01 09:35:09 -08:00

gptq

[Core] Set linear_weights directly on the layer (#3977)

2024-04-11 16:35:51 -04:00

marlin

[Bugfix] Fix marlin kernel crash on H100 (#4218)

2024-04-24 10:35:01 -07:00

squeezellm

Enable CUDA graph for GPTQ & SqueezeLLM (#2318)

2024-01-03 09:52:29 -08:00