周周周 3cc09418f1 support dsv3 use flashmla (#6593)

2026-03-03 11:09:43 +08:00

..

ops

seq_lens related tensor shape -> [max_num_seqs] (#6535)

2026-03-02 11:18:30 +08:00

__init__.py

[XPU] move xpu_attn_backend.py to FastDeploy/fastdeploy/model_executor/layers/backends/xpu (#5878)

2026-01-09 16:34:57 +08:00

append_attn_backend.py

[Feature]Supports SWA based on appendattn (#6547)

2026-03-01 19:02:08 +08:00

attention_selecter.py

…

attention.py

Support Norm before Rope (#6332)

2026-02-05 15:28:52 +08:00

base_attention_backend.py

…

block_multihead_attn_backend.py

[Feature]Support reorder ids to split prefill and decodes (#5779)

2026-02-03 00:28:02 -08:00

flash_attn_backend.py

[BugFix] lazy enable_torch_proxy for cutlass (#6523)

2026-03-02 10:43:58 +08:00

flash_mask_attn_backend.py

[MTP] refactor MTP pre_process (#6358)

2026-02-09 10:47:15 +08:00

iluvatar_attn_backend.py

[Iluvatar] Support CudaGraph and optimize flash_attn_unpadded and fused_neox_rope_embedding (#6553)

2026-03-02 14:07:17 +08:00

mla_attention_backend.py

support dsv3 use flashmla (#6593)

2026-03-03 11:09:43 +08:00

moba_attention_backend.py

[Feature]Support reorder ids to split prefill and decodes (#5779)

2026-02-03 00:28:02 -08:00

native_paddle_backend.py

…

utils.py

…