kangwangamdandYC Yen-Ching Tseng 9f8e916131 [diffusion] post_training: run weight update under torch.inference_mode() (#31263) ...

Co-authored-by: YC Yen-Ching Tseng <yctseng@amd.com>

2026-07-20 11:14:13 +08:00

..

breakable_cuda_graph

Move breakable CUDA graph back into model_executor/runner_backend_utils (#30586)

2026-07-09 14:19:16 +08:00

cache

[diffusion][cache-dit] add dual-transformer Cache-DiT adapter specs (#30150)

2026-07-07 08:10:52 -07:00

disaggregation

[diffusion] rl: support rl rollout for the wan pipeline via a per-request scheduler switch (#30036)

2026-07-15 22:30:51 +08:00

distributed

model: support Pi0.5 (#30633)

2026-07-11 07:50:58 +08:00

entrypoints

Fix image URL response for multiple outputs (#30621)

2026-07-15 14:49:04 +03:00

layers

Remove QServe and FBGEMM FP8 quantization (#31109)

2026-07-17 17:10:34 -07:00

loader

[Diffusion] Use SGLang server for ERNIE-Image prompt enhancement (#31354)

2026-07-18 06:55:28 +03:00

managers

model: support Pi0.5 (#30633)

2026-07-11 07:50:58 +08:00

models

[Multimodal] Support n>1 outputs for GLM-Image generation (#31027)

2026-07-15 20:55:19 +03:00

pipelines

[diffusion] post_training: Add LoRA IPC weight sync via lora_merge mode (#31029)

2026-07-15 08:42:54 +08:00

pipelines_core

[diffusion] Opt in Qwen and Wan multi-output conditioning expansion (#31233)

2026-07-17 12:00:23 -07:00

platforms

Refactor FP4 quantization and remove deprecated JIT kernels (#30448)

2026-07-14 09:22:07 +08:00

post_training

[diffusion] post_training: run weight update under torch.inference_mode() (#31263)

2026-07-20 11:14:13 +08:00

postprocess

[diffusion] improve: improve realtime webui playback pacing (#27148)

2026-06-04 00:33:56 +08:00

realtime

[diffusion] refactor: refactor realtime control state and adapters (#27698)

2026-06-10 12:27:16 +08:00

server_args

[Diffusion] Use SGLang server for ERNIE-Image prompt enhancement (#31354)

2026-07-18 06:55:28 +03:00

utils

[diffusion] Opt in Qwen and Wan multi-output conditioning expansion (#31233)

2026-07-17 12:00:23 -07:00

vla

model: support Pi0.5 (#30633)

2026-07-11 07:50:58 +08:00

ipc_array.py

[diffusion] optimize: optimize frame returns path (#24616)

2026-05-08 12:10:09 +08:00

launch_server.py

model: support Pi0.5 (#30633)

2026-07-11 07:50:58 +08:00

scheduler_client.py

[Diffusion] Diffusion model support log-requests (#23049)

2026-07-05 12:01:26 +03:00

server_warmup.py

model: support Pi0.5 (#30633)

2026-07-11 07:50:58 +08:00

warmup_request_builder.py

model: support Pi0.5 (#30633)

2026-07-11 07:50:58 +08:00