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

..

bagel

fix(bagel): correct CFG position ID concatenation for multimodal RoPE (#5775)

2026-08-05 01:53:59 +08:00

boogu_image

[Model] Support Boogu/Boogu-Image-0.1-Base and Boogu/Boogu-Image-0.1-Edit (#4995)

2026-07-21 18:58:56 +00:00

cosmos3

Cosmos3 logging and prompt improvements (#6325)

2026-08-20 12:04:49 -07:00

cosyvoice3_audio

[Hardware] Enable CosyVoice3 on Moore Threads MUSA (#5164)

2026-07-29 02:46:27 +00:00

diffusers_adapter

[Model] Add native SANA-Video 2B T2V and I2V support (#5508)

2026-08-25 09:46:02 +08:00

dmd2

[Refactor] Unify the LTX-2 and LTX-2.3 pipeline runtime (#5147)

2026-07-20 20:02:30 +08:00

dreamzero

[Core] Add realtime AR-Diffusion tick sessions for LingBot World 2.0 (#5491)

2026-08-16 09:11:20 +08:00

ernie_image

[Refactor][1/N]diffusion/cache: cache-dit (#5226)

2026-07-24 22:42:45 +08:00

flux

diffusion: Fix silent override of explicit guidance_scale=0 (#4999)

2026-08-01 03:19:21 +00:00

flux2

[Diffusion] Make Flux2Pipeline text_encoder_out_layers configurable (#6390)

2026-08-28 09:05:41 -06:00

flux2_klein

[Diffusion] Add FLUX.2-klein BF16 HWR contract (#6651)

2026-08-27 12:51:15 +08:00

glm_image

diffusion: Fix silent override of explicit guidance_scale=0 (#4999)

2026-08-01 03:19:21 +00:00

gr00t

[Rebase] Rebase to vllm v0.25.0 (#5042)

2026-07-12 23:39:53 +08:00

helios

[Frontend] Add midway prompt update for streaming video generation (#4652)

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

hidream_image

[Core][Frontend] Support request-level batching for diffusion pipelines (#4079)

2026-06-29 18:09:19 +02:00

hidream_o1_image

[Model] Add HiDream-O1-Image support (#5194)

2026-08-21 14:15:37 +08:00

hunyuan_image3

[Diffusion] Add native paged KV backend for diffusion workers (#6102)

2026-08-22 21:45:45 +08:00

hunyuan_video

[Perf][Flux2][HunyuanVideo1.5] Skip attention-mask to avoid varlen path (#4645)

2026-08-11 15:31:02 +08:00

internvla_a1

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

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

krea2

[Model] Add Krea 2 text-to-image diffusion model (#4730)

2026-07-11 16:25:41 -07:00

lance

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

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

lingbot_video

[Rebase] Rebase to vllm 0.27.0 (#5976)

2026-08-11 15:23:46 +00:00

lingbot_world

[Core] Add realtime AR-Diffusion tick sessions for LingBot World 2.0 (#5491)

2026-08-16 09:11:20 +08:00

longcat_image

[Bugfix] Honor height/width overrides in LongCatImageEditPipeline (#6222)

2026-08-25 02:09:00 +00:00

longcat_video

[Model] Add LongCat-Video-Avatar-1.5 ai2v, at2v support (#4099)

2026-08-11 15:17:17 +08:00

ltx2

[Diffusion] Add LTX-2.5 Diffusion VAE decoder support (#6189)

2026-08-27 23:54:40 +08:00

mammoth_moda2

[Fix] Restore MammothModa2 support (#6694)

2026-08-28 08:32:32 +08:00

ming_flash_omni

[Bugfix] Adapt ZImagePipeline.forward calling in MingImagePipeline with DiffusionRequestBatch (#4866)

2026-07-04 03:45:49 +00:00

minimax_h3

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

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

mistral_encoder

[Quantization] Add FLUX.2-dev Mistral TE online FP8 support (#5136)

2026-07-20 16:16:31 +08:00

nextstep_1_1

[Core][Frontend] Support request-level batching for diffusion pipelines (#4079)

2026-06-29 18:09:19 +02:00

omnigen2

[Diffusion][Feature] Add SP for Omnigen2 (#3206)

2026-07-08 10:29:41 +08:00

omnivoice

[Core][Frontend] Support request-level batching for diffusion pipelines (#4079)

2026-06-29 18:09:19 +02:00

ovis_image

[Chore]: refactor out unused/redundant params in diffusion pipelines (#1235)

2026-07-02 16:32:05 +02:00

pi0

[Bugfix] Avoid eager pi0 runtime import in pipeline registry (#6322)

2026-08-18 23:42:03 +08:00

qwen_image

[Bugfix][MUSA][Qwen Image] Avoid complex RoPE alias guards (#6110)

2026-08-14 15:51:23 +08:00

sana_video

[Model] Add native SANA-Video 2B T2V and I2V support (#5508)

2026-08-25 09:46:02 +08:00

sana_wm

[Model] Add SANA-WM support(Stage-1 only) (#4061)

2026-08-18 23:57:55 -06:00

schedulers

[BugFix][Diffusion] Add CPU LAPACK fallback for FlowUniPC (#5329)

2026-07-30 15:09:48 +00:00

sd3

[Refactor][1/N]diffusion/cache: cache-dit (#5226)

2026-07-24 22:42:45 +08:00

sdxl

[XPU][SDXL] Fix text encoder input device under CPU offload (#6125)

2026-08-18 14:03:31 +00:00

sensenova_u1

[Bugfix | Model] Fix SenseNova & Use Well-defined Model Configs (#5877)

2026-08-18 16:43:05 +00:00

stable_audio

[CI] Remove AudioX and MammothModa2 support (#6353)

2026-08-23 04:22:05 +00:00

t5_encoder

[Feature]: Migrate magi human example (#4572)

2026-06-22 15:17:00 +08:00

wan2_2

[FEAT]Add FastVideo VSA backend for Wan2.2 (#4820)

2026-08-24 14:48:31 +08:00

z_image

[Chore]: refactor out unused/redundant params in diffusion pipelines (#1235)

2026-07-02 16:32:05 +02:00

__init__.py

[Bugfix] Remove mandatory flash-attn dependency and optimzie docs (#113)

2025-11-30 18:20:44 +08:00

host_weight_contract.py

[Diffusion] Add final-layout BF16 Host Weight Runtime artifacts (#6445)

2026-08-22 21:37:20 +08:00

interface.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

pi0_pipeline_config.py

[Bugfix] Avoid eager pi0 runtime import in pipeline registry (#6322)

2026-08-18 23:42:03 +08:00

progress_bar.py

[UX] Add progress bar for diffusion models (#1652)

2026-03-05 23:56:14 +08:00

utils.py

[Quantization] Redo Z-Image text encoder FP8 online quantization (#3279)

2026-05-06 16:37:36 +00:00