SYLARandlishunyang12 507cb1d834 [Attention] [P1] Keep H3 VSA policy in the model and share sparse primitives (#7535) ...

Signed-off-by: lishunyang12 <lishunyang12@users.noreply.github.com>
Co-authored-by: lishunyang12 <lishunyang12@users.noreply.github.com>

2026-09-16 15:47:31 +08:00

..

attention_backends

[Attention] [P1] Keep H3 VSA policy in the model and share sparse primitives (#7535)

2026-09-16 15:47:31 +08:00

cache_acceleration

[Model][Feat] Support Minimax-H3 quality grading requests through dynamic loading/unloading with Cache-DiT (#5853)

2026-08-07 09:07:23 +00:00

offloader

[diffusion][feature] Add component-selective offload policies (#5929)

2026-09-06 05:40:09 +08:00

parallelism

[Bugfix] Add explicit error when using CFGP with distilled Cosmos3 models (#7427)

2026-09-11 22:59:49 +00:00

attention_backends.md

[Feature][MiniMax-H3] Complete VSA and Ulysses support (#6909)

2026-09-02 07:19:21 -04:00

cpu_offload.md

[diffusion][feature] Add component-selective offload policies (#5929)

2026-09-06 05:40:09 +08:00

device_side_video_postprocess.md

[Core][Diffusion] Add a typed pre-D2H video media contract (#6615)

2026-09-10 09:50:54 +00:00

execution_modes.md

[Core][Diffusion] Refactor prompt updates into a modality-ready interaction framework (#6294)

2026-09-07 11:40:04 +08:00

frame_interpolation.md

[REFACT]Refactor diffusion outputs to payload metadata (#4922)

2026-07-14 17:24:03 +08:00

lora.md

[Feature][MiniMax-H3] Support the full LightX2V Turbo LoRA matrix (#7062)

2026-09-06 20:23:36 +08:00

mot_config.md

[Enhancement] Introduce High-Performance MoT (Mixture-of-Tokens) Kernels: Triton Implementation (#3960)

2026-06-16 03:08:02 +00:00

regional_compilation.md

[Model][MiniMax-H3] Multi-GPU serving fixes and layer-wise offload component selection (#7047)

2026-09-14 10:02:30 -07:00

startup_and_loading.md

docs: align User Guide feature taxonomy (#6045)

2026-08-12 00:20:56 +08:00