Yongye ZhuandRoger Wang 0d2732dd91 [MLA Attention Backend] Add TOKENSPEED_MLA backend for DSR1/Kimi K25 prefill + decode on Blackwell (#41778) ...

Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>

2026-05-13 23:48:02 -07:00

..

__init__.py

[5/N][Attention] Finish eliminating vllm/attention folder (#32064)

2026-01-27 10:02:51 -05:00

attention.py

expose flex block size for batch invariant mode (#41252)

2026-05-13 14:11:57 -07:00

chunked_local_attention.py

[MyPy] Enable mypy for vllm/model_executor/layers/ (#40159)

2026-04-21 20:15:02 -07:00

cross_attention.py

[Core] Avoid seq_lens_cpu GPU->CPU sync (#40654)

2026-04-24 00:35:55 +00:00

encoder_only_attention.py

[MyPy] Enable mypy for vllm/model_executor/layers/ (#40159)

2026-04-21 20:15:02 -07:00

kv_transfer_utils.py

[compile] Allow strings in custom ops without regressing compilation times (#38123)

2026-04-10 07:26:37 +00:00

mla_attention.py

[MLA Attention Backend] Add TOKENSPEED_MLA backend for DSR1/Kimi K25 prefill + decode on Blackwell (#41778)

2026-05-13 23:48:02 -07:00

mm_encoder_attention.py

[Perf] FP8 FlashInfer Attn for ViT (#38065)

2026-04-27 13:44:15 +08:00

static_sink_attention.py

[compile] Allow strings in custom ops without regressing compilation times (#38123)

2026-04-10 07:26:37 +00:00