Lianmin Zheng 9d5fa68b90 Use torch.compile to fuse flash attention decode metadata preparation (#6973)

2025-06-08 23:05:40 -07:00

..

triton_ops

Support sliding window in triton backend (#6509)

2025-05-30 01:11:53 -07:00

aiter_backend.py

AITER backend extension and workload optimizations (#6838)

2025-06-05 23:00:18 -07:00

base_attn_backend.py

Add intel_amx backend for Radix Attention for CPU (#6408)

2025-05-30 21:37:42 -07:00

cutlass_mla_backend.py

Cutlass MLA decode - fix dtype error (#5868)

2025-04-28 21:12:58 -07:00

double_sparsity_backend.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

flashattention_backend.py

Use torch.compile to fuse flash attention decode metadata preparation (#6973)

2025-06-08 23:05:40 -07:00

flashinfer_backend.py

[Feature] Support Flashinfer fmha on Blackwell (#6930)

2025-06-06 12:57:50 -07:00

flashinfer_mla_backend.py

[Feature] Support Flashinfer fmha on Blackwell (#6930)

2025-06-06 12:57:50 -07:00

flashmla_backend.py

[Feat] Support FlashMLA backend with MTP and FP8 KV cache (#6109)

2025-05-15 00:48:09 -07:00

intel_amx_backend.py

Add intel_amx backend for Radix Attention for CPU (#6408)

2025-05-30 21:37:42 -07:00

merge_state.py

feat: Add a unified merge_state API (#5428)

2025-05-05 10:32:33 -07:00

tbo_backend.py

Minor refactor two-batch overlap (#6682)

2025-05-28 15:54:17 -07:00

torch_native_backend.py

Add intel_amx backend for Radix Attention for CPU (#6408)

2025-05-30 21:37:42 -07:00

triton_backend.py

Support sliding window in triton backend (#6509)

2025-05-30 01:11:53 -07:00

utils.py

[Feat] Support FlashMLA backend with MTP and FP8 KV cache (#6109)

2025-05-15 00:48:09 -07:00

vision.py

refactor: minor refactors regarding multimodal processing (#6187)

2025-05-17 22:53:20 -07:00