hurukawaandHongsheng Liu 855719e331 [Bugfix] Release GPU memory after diffusion execution failures (#6385) ...

Signed-off-by: nagisa <1434936049@qq.com>
Signed-off-by: nagisa-kun <1434936049@qq.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>

2026-08-22 18:32:48 +00:00

..

assets

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

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

attention

[Rebase] Rebase to vllm 0.26.0 (#5443)

2026-07-27 23:35:49 +08:00

benchmarks

[Model][Frontend] MiniMax-H3: Add opt-in planar video response encoding (#6288)

2026-08-22 04:04:14 +00:00

buildkite

[CI] Expand E2E source_file_dependencies for shared model code (#5994)

2026-08-15 12:01:20 +08:00

config

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

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

core

[Bugfix][MiniCPM-o] Fix async-chunk snapshot replacement and prompt cleanup (#6406)

2026-08-21 16:37:21 +08:00

cosmos3

[Feat] Cosmos3 session-memory port for UND text K/V (RFC #4480 Phase 0) (#4657)

2026-08-09 20:53:17 +08:00

dfx

[CI/Build] Update perf baselines base on 8/1-8/7 7 days avg (#6201)

2026-08-20 18:56:23 +00:00

diffusion

[Bugfix] Release GPU memory after diffusion execution failures (#6385)

2026-08-22 18:32:48 +00:00

distributed

[Bugfix][MiniCPM-o] Fix async-chunk snapshot replacement and prompt cleanup (#6406)

2026-08-21 16:37:21 +08:00

docs

[Bugfix] Stop reading the removed OmniRequestOutput.request_output accessor (#6172)

2026-08-15 07:07:28 +08:00

e2e

[Core] Add UniProc diffusion executor for single-GPU (#6308)

2026-08-22 21:58:21 +08:00

engine

[Bugfix][Core] Fix replica device split when a stage omits devices (#5445)

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

entrypoints

[Bugfix] Preserve pipeline sampling constraints (#6182)

2026-08-22 14:57:09 +00:00

examples

[Tests] Re-enable previously skipped e2e/example tests (#5641)

2026-08-21 14:45:58 +08:00

helpers

[Bugfix] Preserve prompt token usage details (#5181)

2026-08-20 20:05:05 +00:00

host_weight_runtime

[Core] Add explicit post-load Host Weight Runtime publication (#6427)

2026-08-21 14:26:49 +08:00

inputs

[Model][Feat] Support Minimax-H3 quality grading requests through dynamic loading/unloading with Cache-DiT (#5853)

2026-08-07 09:07:23 +00:00

metrics

[Metrics] Add Diffusion Metrics (#4755)

2026-08-13 06:59:13 +00:00

model_executor

[Bugfix][MiniCPM-o] Cap offline Talker generation at remaining context (#6458)

2026-08-22 12:42:49 +08:00

model_extras

Cosmos3 transfer fix (#5614)

2026-08-20 11:40:51 -07:00

model_tests

[Bugfix] Fix LongCat TeaCache CFG negative-branch guidance kwargs (#6181)

2026-08-22 01:22:26 +00:00

platforms

[Bugfix][MiniCPM-o] Fix async-chunk snapshot replacement and prompt cleanup (#6406)

2026-08-21 16:37:21 +08:00

profile

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

quantization

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

reasoning

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

tools

[Migrate] Move sampling parameter overrides from legacy dispatches to TTS adapters (#5272)

2026-08-16 21:14:26 -07:00

utils

[Feature] timestamps for TTS via a forced-aligner pooling (#4795)

2026-08-12 05:33:34 +00:00

worker

[Bugfix][MiniCPM-o] Fix async-chunk snapshot replacement and prompt cleanup (#6406)

2026-08-21 16:37:21 +08:00

__init__.py

[Fix] Resolve comments & update vLLM-Omni name usages. (#122)

2025-12-01 08:05:24 +08:00

conftest.py

[Tests] Run diffusion tiny model tests in parallel (#6339)

2026-08-20 12:46:32 -04:00

test_indextts2_optional_dependencies.py

[New Model] Support IndexTTS 2.5 (#5957)

2026-08-11 23:00:35 +00:00