Alex Brooksand汪志鹏 7ff11d8d0f [Perf/Fix] Reimplement Batched CFG Forward for Bagel (#4098) ...

Signed-off-by: Alex Brooks <albrooks@redhat.com>
Co-authored-by: 汪志鹏 <wangzhipeng628@gmail.com>

2026-06-22 07:37:21 +08:00

..

attention

[XPU] Add sage_attn backend (#3785)

2026-06-09 19:28:18 +08:00

cache

[Cache Refactor 1/N] Simplify CacheDiT Integration (#2527)

2026-06-20 08:30:44 +08:00

distributed

[Bugfix] Fix HSDP + FP8 online quantization compatibility (#4494)

2026-06-19 15:06:27 +08:00

executor

[Bugfix] Surface all-rank diffusion RPC failures (#4403)

2026-06-13 15:42:16 +08:00

hooks

[Feat] Support MagCache (#1287)

2026-05-29 22:26:23 +08:00

layers

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

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

lora

[Feature] LoRA support for SenseNova-U1 (#3971)

2026-06-09 14:36:35 +00:00

model_loader

[Bugfix] Fix HSDP + FP8 online quantization compatibility (#4494)

2026-06-19 15:06:27 +08:00

models

[Perf/Fix] Reimplement Batched CFG Forward for Bagel (#4098)

2026-06-22 07:37:21 +08:00

offloader

[Refactor] Migrate existing pipelines to use SupportsComponentDiscovery for offload discovery (#3076)

2026-06-17 21:33:06 +08:00

postprocess

[Bugfix] Fix RIFE device selection for CPU-transported videos (#2876)

2026-04-17 19:19:34 +08:00

profiler

[Feature] support sp for hunyuan (#2163)

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

quantization

[Bugfix] Fix HSDP + FP8 online quantization compatibility (#4494)

2026-06-19 15:06:27 +08:00

sched

[Diffusion] Unify diffusion request identity on request_id (#3744)

2026-05-28 14:56:56 +08:00

utils

[Perf/Fix] Reimplement Batched CFG Forward for Bagel (#4098)

2026-06-22 07:37:21 +08:00

worker

[ROCm] [CI] Fix diffusion test (#4574)

2026-06-20 08:31:39 +08:00

__init__.py

[Diffusion] Qwen image support (#82)

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

compile.py

[Perf]Enable regional compile for Cosmos3 HSDP blocks (#4485)

2026-06-18 16:54:05 +08:00

config.py

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

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

data.py

[Diffusion] add GR00T-N1.7 pipeline with OpenPI serving (#3798)

2026-06-20 11:45:26 +08:00

diffusion_engine.py

[Refactor]Base class for dit pipelines with unified parameter declaration[1/N] (#4225)

2026-06-18 17:14:18 +08:00

envs.py

[Refactor] Extract has_flash_attn_pkg in OmniPlatform (#3068)

2026-04-27 04:07:34 +00:00

forward_context.py

[Model] Add Ming-flash-omni-2.0 Image Generation (Diffusion) Stage (#2875)

2026-05-27 23:21:20 +08:00

inline_stage_diffusion_client.py

[refactor] Refactor guardrail error handling - add 400 error code (#4297)

2026-06-11 21:09:24 +00:00

io_support.py

[Refactor]Base class for dit pipelines with unified parameter declaration[1/N] (#4225)

2026-06-18 17:14:18 +08:00

ipc.py

[Bugfix] Surface all-rank diffusion RPC failures (#4403)

2026-06-13 15:42:16 +08:00

model_metadata.py

[Feature] HunyuanImage-3.0 IT2I: multi-image input + prompt API cleanup (#3444)

2026-05-14 16:58:40 +08:00

output_formatter.py

[Refactor] Extract diffusion output formatting boundary (#4407)

2026-06-15 22:50:50 +08:00

registry.py

[Diffusion] add GR00T-N1.7 pipeline with OpenPI serving (#3798)

2026-06-20 11:45:26 +08:00

request.py

[Diffusion] Unify diffusion request identity on request_id (#3744)

2026-05-28 14:56:56 +08:00

stage_diffusion_client.py

[BugFix]Fix diffusion profile RPC None results (#4372)

2026-06-21 17:14:21 +08:00

stage_diffusion_proc.py

[Refactor] Omni Stage Runtime and Distributed Replica Control Plane (#3855)

2026-06-17 14:13:20 +08:00