| .. |
|
__init__.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
allclose_default.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
conftest.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
quant_utils.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
test_activation.py
|
172dee2573
(2/N) Triton Backend: integrate Triton activation kernels (#1126)
|
пре 1 недеља |
|
test_attention.py
|
61aed092a5
rocm: add support for FP8 KV cache in the custom paged attention kkernels (#1066)
|
пре 1 месец |
|
test_attention_selector.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
test_awq_triton.py
|
fcfcfc65e1
quants: add triton kernels for AWQ (#946)
|
пре 1 месец |
|
test_blocksparse_attention.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
test_cache.py
|
a113309876
kernel: add meta functions for ops to prevent graph breaks (#1019)
|
пре 1 месец |
|
test_causal_conv1d.py
|
9bdf8d5bfa
mamba: enable continuous batching for mamba kernels (#1055)
|
пре 1 месец |
|
test_cutlass.py
|
a113309876
kernel: add meta functions for ops to prevent graph breaks (#1019)
|
пре 1 месец |
|
test_encoder_decoder_attn.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
test_flash_attn.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
test_flashinfer.py
|
a62f0925fe
update flashinfer test (#945)
|
пре 1 месец |
|
test_fp8_quant.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
test_int8_quant.py
|
8976805f90
kernel: asymmetric AQ AZP quantization kernels (#1048)
|
пре 1 месец |
|
test_layernorm.py
|
0c17153073
(1/N) Triton Backend: integrate Triton layernorm kernels (#1125)
|
пре 1 недеља |
|
test_machete_gemm.py
|
a113309876
kernel: add meta functions for ops to prevent graph breaks (#1019)
|
пре 1 месец |
|
test_mamba_ssm.py
|
9bdf8d5bfa
mamba: enable continuous batching for mamba kernels (#1055)
|
пре 1 месец |
|
test_marlin_gemm.py
|
a113309876
kernel: add meta functions for ops to prevent graph breaks (#1019)
|
пре 1 месец |
|
test_moe.py
|
fa84f8102e
kernels: split marlin kernels for faster compile, fix MoE, temporarily remove HQQ (#1119)
|
пре 1 недеља |
|
test_permute_cols.py
|
93bc863591
feat: Machete Kernels for Hopper GPUs (#842)
|
пре 2 месеци |
|
test_pos_encoding.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
test_prefix_prefill.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
пре 2 месеци |
|
utils.py
|
7b6501bd05
tests: refactor model tests (#1078)
|
пре 1 месец |
AlpinDale
172dee2573
(2/N) Triton Backend: integrate Triton activation kernels (#1126)