3 people cb6a8739c1 [Feature] Add trajectory recording to BAGEL denoising loop (#2483) ...

Signed-off-by: Zhengyuan Su <su.zhengyuan@u.nus.edu>
Signed-off-by: Zhengyuan Su (苏政渊) <su.zhengyuan@u.nus.edu>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: Samit <285365963@qq.com>

2026-04-08 10:43:01 +08:00

..

attention

[Rebase] Rebase to vllm v0.19.0 (#2475)

2026-04-04 15:03:32 +08:00

cache

[Model] Add two stages inference for model LTX-2 distilled. (#2260)

2026-04-03 23:17:49 +08:00

distributed

[Feature] Support vae tiling parallel encode (#2368)

2026-04-06 14:47:45 +00:00

executor

[Feat] Support step-boundary abort in diffusion (#1769)

2026-04-01 17:25:02 +08:00

hooks

[Feature] support sp for hunyuan (#2163)

2026-03-27 17:28:13 +08:00

layers

[AutoRound] Add offline quantized W4A16 model support (#1777)

2026-04-02 13:57:53 +08:00

lora

[Feature] Enable LoRA adapter injection for BAGEL (#2490)

2026-04-06 19:30:52 +08:00

model_loader

[AutoRound] Add offline quantized W4A16 model support (#1777)

2026-04-02 13:57:53 +08:00

models

[Feature] Add trajectory recording to BAGEL denoising loop (#2483)

2026-04-08 10:43:01 +08:00

offloader

[Feat] Enable Layerwise CPU offloading for SD3.5, Ovis-Image, Nextstep_1, LongCat-Image (#2339)

2026-04-08 09:55:09 +08:00

profiler

[Feature] support sp for hunyuan (#2163)

2026-03-27 17:28:13 +08:00

sched

[Feat] Support step-boundary abort in diffusion (#1769)

2026-04-01 17:25:02 +08:00

utils

[daVinci-MagiHuman][Doc][BugFix] Update model support for daVici-MagiHuman and fix media utils bug (#2542)

2026-04-07 17:16:12 +08:00

worker

[Rebase] Rebase to vllm v0.19.0 (#2475)

2026-04-04 15:03:32 +08:00

__init__.py

[Diffusion] Qwen image support (#82)

2025-11-28 14:22:11 +08:00

compile.py

[core] add torch compile for diffusion (#684)

2026-01-09 04:05:44 +13:00

data.py

[Feature] Add trajectory recording to BAGEL denoising loop (#2483)

2026-04-08 10:43:01 +08:00

diffusion_engine.py

[Feature] Add trajectory recording to BAGEL denoising loop (#2483)

2026-04-08 10:43:01 +08:00

envs.py

[Frontend] Rewrite video API for async job lifecycle (#1665)

2026-03-13 17:11:06 +08:00

forward_context.py

[Feature]: add Ulysses advanced_uaa mode (#1379)

2026-03-24 14:55:54 +08:00

ipc.py

[Feature] Add trajectory recording to BAGEL denoising loop (#2483)

2026-04-08 10:43:01 +08:00

registry.py

[Model Support]: Magihuman support (#2301)

2026-04-04 09:29:09 +00:00

request.py

[Bug Fix] Resolve broken image issue when TP is enabled and no seed is provided. (#2176)

2026-03-26 16:56:46 +00:00

stage_diffusion_client.py

[Rebase] Rebase to vllm v0.19.0 (#2475)

2026-04-04 15:03:32 +08:00

stage_diffusion_proc.py

[Feature] Add trajectory recording to BAGEL denoising loop (#2483)

2026-04-08 10:43:01 +08:00