Aryan Kumar 47bb2539f8 [feat]: NVFP4 FA4 deploy path applies exact Q/K preconditioning and per-block BF16 attention opt-out

2026-11-01 02:22:35 -08:00

..

basic

[feat]: NVFP4 FA4 deploy path applies exact Q/K preconditioning and per-block BF16 attention opt-out

2026-11-01 02:22:35 -08:00

eval

[feat] Add end-to-end MMAudio V2A training, inference, and evaluation (#1648)

2026-10-07 04:25:00 -07:00

gradio

[feat] Add MiniMax H3 LoRA inference and preview launchers (#1771)

2026-08-27 14:43:39 -07:00

lora

[refactor] VSA: rename the sm100a block-sparse kernel to ptx_blackwell (#1956)

2026-10-17 21:10:19 -07:00

ltx2_3

[misc] yapf pass under CI's interpreter (3.12) + pin hook language_version (#1702)

2026-08-08 22:24:17 -07:00

optimizations

[misc] Clean up QAD 5090 example inference scripts (#1496)

2026-10-07 03:44:49 -07:00

sta_mask_search

[Misc] Remove STA (#1124)

2026-02-23 15:14:42 -08:00