jennyyyyzhen 2aab2bb543 [ROCM] Optimize ROCM_AITER_FA spec decode eagle performance (#34541) ...

Signed-off-by: jennyyyyzhen <yzhen@hmc.edu>

2026-02-20 20:32:05 -08:00

..

mla

[Bugfix][MTP][Sparse MLA] Allow sparse MLA with MTP to run with FULL cudagraphs (#34457)

2026-02-17 14:01:27 -05:00

__init__.py

[V1] Implement vLLM V1 [1/N] (#9289)

2024-10-22 01:24:07 -07:00

cpu_attn.py

[cpu][performance] CPU Paged Attention NEON BFMMLA BF16 Implementation (#32263)

2026-02-06 15:01:48 +08:00

fa_utils.py

[XPU][9/N] clean up existing ipex code/doc (#34111)

2026-02-11 00:27:15 -08:00

flash_attn_diffkv.py

[1/N][Attention] Restructure attention: move files (#31916)

2026-01-09 13:10:24 -08:00

flash_attn.py

Reapply [Attention][FA3] Update FA3 to include new swizzle optimization (#34043)

2026-02-11 07:07:56 -08:00

flashinfer.py

[Revert] Fix performance regression for GLM-4.7-GPTQ decode and MTP acceptance rate (#33771)

2026-02-08 08:13:24 -08:00

flex_attention.py

[Bugfix][Attention] Explicitly report support for kv_cache_dtype bfloat16 (#32795)

2026-01-22 19:05:18 +00:00

gdn_attn.py

[GDN] Use CPU tensors to build GDN metadata (#34498)

2026-02-13 01:24:45 -08:00

linear_attn.py

[V1][Hybrid] Mamba Prefix Caching with align mode (#30877)

2026-01-23 09:56:48 -08:00

mamba1_attn.py

[Misc] Add get_name to missing AttentionBackends (#32698)

2026-01-23 10:35:44 +00:00

mamba2_attn.py

[Misc] offest -> offset in comments and variable names (#33444)

2026-01-30 20:19:22 -08:00

mamba_attn.py

[V1][Hybrid] Mamba Prefix Caching with align mode (#30877)

2026-01-23 09:56:48 -08:00

registry.py

[Attention] Add FlashInfer Sparse MLA backend (#33451)

2026-02-12 17:21:54 +00:00

rocm_aiter_fa.py

[ROCM] Optimize ROCM_AITER_FA spec decode eagle performance (#34541)

2026-02-20 20:32:05 -08:00

rocm_aiter_unified_attn.py

move checks out of unified_kv_cache_update custom op (#33943)

2026-02-07 05:30:09 -08:00

rocm_attn.py

move checks out of unified_kv_cache_update custom op (#33943)

2026-02-07 05:30:09 -08:00

short_conv_attn.py

[Misc] Add get_name to missing AttentionBackends (#32698)

2026-01-23 10:35:44 +00:00

tree_attn.py

[Misc] Replace Optional[X] with X | None syntax (#33332)

2026-01-30 01:56:59 -08:00

triton_attn.py

move checks out of unified_kv_cache_update custom op (#33943)

2026-02-07 05:30:09 -08:00

utils.py

[perf] Avoid dtype promotion sync in mamba_get_block_table_tensor (#34870)

2026-02-20 06:21:56 -08:00