Aryan KumarandAryan Kumar 4b4df1a94e [perf] Fused Q/K preconditioning + NVFP4 quantize for FA4 NVFP4 attention (bit-identical) (#2052) ...

Co-authored-by: Aryan Kumar <aryan5v@users.noreply.github.com>

2026-11-03 01:07:11 -08:00

..

cudnn_sdpa.py

[perf] MiniMax H3: faster FP4 attention padding, uint8 VAE pixel path and cuDNN FP4 GEMMs (#2046)

2026-11-01 22:23:10 -08:00

fa4_env_cache.py

[bugfix]: get the full CI suite green on main (pytest 9 flag teardown, Qwen golden env) (#2059)

2026-11-02 23:32:40 -08:00

flash_attn_cute.py

[bugfix] FA4 JIT cache vs tuning env, MXFP8 import and scale dtype, H3 VSA gate double quantization (#2049)

2026-11-01 19:11:25 -08:00

flash_attn_default.py

[feat] Add a typed environment-variable registry, policy doc, and contract test (#1896)

2026-09-29 12:50:45 -07:00

flash_attn_no_pad.py

[feat] Add a typed environment-variable registry, policy doc, and contract test (#1896)

2026-09-29 12:50:45 -07:00

nvfp4_fused_qk_quant.py

[perf] Fused Q/K preconditioning + NVFP4 quantize for FA4 NVFP4 attention (bit-identical) (#2052)

2026-11-03 01:07:11 -08:00

nvfp4_qk_precondition.py

[feat] Pika H3 Flex NVFP4 QAD training: calibration, QAD, FP4 attention and export (#2051)

2026-11-03 01:04:26 -08:00