Lucas Wilkinson d200972e7f

[Bugfix] Marlin 2:4 temp fix for large M dim (>256) (#10464) ...

Signed-off-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>

2024-11-19 19:40:33 -08:00

..

attention

Support Roberta embedding models (#9387)

2024-11-14 21:23:29 +00:00

core

…

cpu

Fix: Build error seen on Power Architecture (#10421)

2024-11-19 09:34:57 -08:00

cutlass_extensions

[Kernel] Initial Machete W4A8 support + Refactors (#9855)

2024-11-18 12:59:29 -07:00

mamba

…

moe

…

prepare_inputs

[Core] Flashinfer - Remove advance step size restriction (#10282)

2024-11-13 16:29:32 +08:00

quantization

[Bugfix] Marlin 2:4 temp fix for large M dim (>256) (#10464)

2024-11-19 19:40:33 -08:00

rocm

…

activation_kernels.cu

…

cache_kernels.cu

…

cache.h

…

cuda_compat.h

…

cuda_utils_kernels.cu

…

cuda_utils.h

…

custom_all_reduce_test.cu

…

custom_all_reduce.cu

…

custom_all_reduce.cuh

…

dispatch_utils.h

…

layernorm_kernels.cu

[torch.compile] Fuse RMSNorm with quant (#9138)

2024-11-08 21:20:08 +00:00

layernorm_quant_kernels.cu

[torch.compile] Fuse RMSNorm with quant (#9138)

2024-11-08 21:20:08 +00:00

ops.h

[torch.compile] Fuse RMSNorm with quant (#9138)

2024-11-08 21:20:08 +00:00

permute_cols.cu

…

pos_encoding_kernels.cu

…

torch_bindings.cpp

[Model][Quantization] HQQ support through Marlin kernel expansion (#9766)

2024-11-19 13:31:12 -08:00

type_convert.cuh

[torch.compile] Fuse RMSNorm with quant (#9138)

2024-11-08 21:20:08 +00:00