Wenjia LiandWeiQing Chen d9980c75e8 Perf/minimax h3 dit swiglu rope cache npu (#6410) ...

Signed-off-by: Wenjia Li <wjialish@gmail.com>
Co-authored-by: WeiQing Chen <40507679+david6666666@users.noreply.github.com>

2026-08-29 09:14:52 +00:00

..

__init__.py

[Diffusion] Add cache-dit and unify diffusion cache backend interface (#250)

2025-12-15 15:10:45 +08:00

test_cache_backends.py

[CI]Remove DreamID-Omni and MagiHuman support (#6357)

2026-08-26 01:31:12 +00:00

test_cache_dit_request_runtime.py

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

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

test_cache_dit.py

[CI]Remove DreamID-Omni and MagiHuman support (#6357)

2026-08-26 01:31:12 +00:00

test_magcache_hook.py

[Bugfix][Diffusion] Apply MagCache residual once per skipped step (#5561)

2026-07-30 07:09:25 -07:00

test_prompt_embed_cache.py

[feat] Support prompt embedding caching for diffusion model (#2962)

2026-05-29 15:36:03 +08:00

test_stepcache.py

[Perf][DreamZero]dreamzero supports CUDA graph, torch.compile, and DiT Caching. (#4213)

2026-06-19 14:24:37 +08:00

test_teacache_extractors.py

Perf/minimax h3 dit swiglu rope cache npu (#6410)

2026-08-29 09:14:52 +00:00

test_teacache.py

[Refactor][OutputProcessor 2/8]: OmniRequestOutput should inherit RequestOutput and no nested wrap-up (#5146)

2026-08-12 22:58:23 +08:00