QianCyrusandHongsheng Liu 8ef77be66e [Model] Avoid repeated attention and MLP work in Klein TP (#8553) ...

Signed-off-by: QianCyrus <101633534+QianCyrus@users.noreply.github.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>

2026-10-07 21:21:05 +08:00

..

attention

[Bugfix] Add TRTLLM attention custom op and execution contract (#7214)

2026-10-04 09:33:51 +00:00

cache

Kandinsky6 (#8537)

2026-10-06 15:55:15 +08:00

diffusion_kv

[Diffusion][HunyuanImage3] Add request paged prefix caching (#7719)

2026-09-23 15:34:56 +00:00

distributed

[Core] Add batch parallel VAE decode (#8162)

2026-10-06 13:51:35 +08:00

executor

[Bugfix] Keep client error status in the single-GPU diffusion executor (#8474)

2026-10-05 13:09:34 -04:00

hooks

[Feature][Diffusion] Add BOOGU Image sequence parallel support (#5718)

2026-09-25 16:18:23 +08:00

interaction

[Core][Diffusion] Camera interaction for diffusion streaming generation (LingBot World 2 as example) (#7198)

2026-09-20 16:19:41 +00:00

layers

[Perf] Fuse AdaLayerNorm CUDA eager path with Triton (#8231)

2026-10-06 15:11:45 +00:00

lora

[Model] Avoid repeated attention and MLP work in Klein TP (#8553)

2026-10-07 21:21:05 +08:00

model_loader

[feature] Add HSDP pre-sharded option (#7948)

2026-10-02 09:35:09 +00:00

models

[Model] Avoid repeated attention and MLP work in Klein TP (#8553)

2026-10-07 21:21:05 +08:00

offloader

[Bugfix][CI/Build] Fix layerwise offload shutdown and enable it for the LTX-2 distilled T2V accuracy test (#8453)

2026-10-04 07:25:51 +00:00

postprocess

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

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

profiler

[Bugfix] Report stepwise Qwen-Image profiler metrics (#4517)

2026-07-03 04:32:22 +00:00

quantization

[Bugfix] Fix HSDP compatibility with the new online FP8 linear method (#5677)

2026-08-14 15:45:55 +08:00

sched

[Diffusion][HunyuanImage3] Add request paged prefix caching (#7719)

2026-09-23 15:34:56 +00:00

utils

Kandinsky6 (#8537)

2026-10-06 15:55:15 +08:00

worker

[Bugfix] Keep client error status in the single-GPU diffusion executor (#8474)

2026-10-05 13:09:34 -04:00

__init__.py

[Diffusion] Qwen image support (#82)

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

cancellation.py

[Bugfix] Stop cancelled MiniMax-H3 requests at denoising boundaries (#7844)

2026-09-25 13:36:09 +08:00

compile.py

fix: keep DLO hooks outside regional compilation (#6073)

2026-08-22 21:44:03 +08:00

config.py

[Diffusion][Attention] Support per-role attention backend via CLI (#2681)

2026-05-10 11:59:05 +08:00

data.py

[Core] Add batch parallel VAE decode (#8162)

2026-10-06 13:51:35 +08:00

diffusion_engine.py

[feature] Enable multi-node inference through Ray backend (#7938)

2026-10-05 02:53:36 +00:00

envs.py

[Kernel][Boogu-Image] Fuse Q/K RMSNorm + interleaved RoPE via fused_qk_norm_rope (#6982)

2026-09-13 09:39:27 +08:00

forward_context.py

[Feature][Diffusion] Add BOOGU Image sequence parallel support (#5718)

2026-09-25 16:18:23 +08:00

inline_stage_diffusion_client.py

[Feature] Transfer AR-to-diffusion payloads over NIXL (#6264)

2026-09-23 18:26:53 +08:00

io_support.py

[Feature] Support text encoder Disaggregation for H3 (#5885)

2026-08-28 14:27:05 +08:00

ipc.py

[Experimental][Perf] Eliminate redundant host copies from generated-video transport (#7355)

2026-09-30 11:37:57 +08:00

media.py

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

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

model_metadata.py

Kandinsky6 (#8537)

2026-10-06 15:55:15 +08:00

output_formatter.py

[Bugfix] Keep async video jobs queued until inference and abort on delete (#6759)

2026-09-15 20:43:30 -07:00

registry.py

Kandinsky6 (#8537)

2026-10-06 15:55:15 +08:00

request.py

[Bugfix] Stop cancelled MiniMax-H3 requests at denoising boundaries (#7844)

2026-09-25 13:36:09 +08:00

stage_diffusion_client.py

[feature] Enable multi-node inference through Ray backend (#7938)

2026-10-05 02:53:36 +00:00

stage_diffusion_proc.py

[feature] Enable multi-node inference through Ray backend (#7938)

2026-10-05 02:53:36 +00:00

vllm_config.py

[Diffusion][HunyuanImage3] Add request paged prefix caching (#7719)

2026-09-23 15:34:56 +00:00