.. |
all_reduce
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
attention
|
0256ed236b
feat: windows support (#790)
|
2 hónapja |
backup
|
f8dfac6372
chore: attention refactor and upstream sync apr01 (#365)
|
9 hónapja |
core
|
0256ed236b
feat: windows support (#790)
|
2 hónapja |
cpu
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
cutlass_extensions
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
1 hónapja |
hadamard
|
5d288aa76c
feat: add fast hadamard transformation kernels (#232)
|
11 hónapja |
mamba
|
0256ed236b
feat: windows support (#790)
|
2 hónapja |
moe
|
0256ed236b
feat: windows support (#790)
|
2 hónapja |
prepare_inputs
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
punica
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
quantization
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
1 hónapja |
sampling
|
bfc8988116
feat: add cuda sampling kernels for top_k and top_p (#828)
|
1 hónapja |
activation_kernels.cu
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
cache.h
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
cache_kernels.cu
|
0256ed236b
feat: windows support (#790)
|
2 hónapja |
cuda_compat.h
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
cuda_utils.h
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
1 hónapja |
cuda_utils_kernels.cu
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
dispatch_utils.h
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
layernorm_kernels.cu
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
ops.h
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
1 hónapja |
permute_cols.cu
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
1 hónapja |
pos_encoding_kernels.cu
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 hónapja |
reduction.cuh
|
0256ed236b
feat: windows support (#790)
|
2 hónapja |
torch_bindings.cpp
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
1 hónapja |