Morph MORPH
SIGN IN SIGN UP
dhruvhead / vllm-omni UNCLAIMED
0 0 7 Python

COMMITS

main
August 26, 2026
Z
[Doc] Add MiMo-Audio recipe for RTX 5090/5090D 32GB (#6559)
Zhichao Zhang committed
816335c
M
[Feature] Support MiniMax-H3 Turbo LoRA with DLO (#6550)
Mu GuanLin committed
59cf3ad
G
Revert "[Bugfix] Restore name-based model detection for HF cache snapshot paths" (#6642)
Gao Han committed
7439ed4
Y
[Bugfix] Restore name-based model detection for HF cache snapshot paths (#6624)
Yueqian Lin committed
7ba5e8a
A
[Model] Optimize CosyVoice3 TensorRT stream handoff (#5673)
Allen Wu committed
f90e992
A
[Bugfix][MiniCPM-o] Serve native duplex from shipping YAMLs and fence… (#6619)
amy-why-3459 committed
5de6f7c
H
text and image metric (#6150)
Honghan Zhu committed
7ae15c7
W
[CI]Remove DreamID-Omni and MagiHuman support (#6357)
wangyu committed
aeb348d
X
[Perf][NPU] Enable native GQA & fused AddNorm for MiniMax-H3 encoder (#6040)
XiongZhongwei committed
40dcfdb
August 25, 2026
H
[Bugfix] Reconcile env-var inventory with post-#6217 main drift (#6631)
Hongsheng Liu committed
a708ae5
H
[Doc] Add environment variables configuration reference (#6217)
Hongsheng Liu committed
780e8d4
S
[CI/Build] Pin GGUF plugin for diffusion nightly tests (#6303)
Satellite committed
6639c82
A
[Bugfix] Handle DoS Overflow Cases (#6598)
Alex Brooks committed
0eb47b7
N
[Benchmark] Add local OmniInteract realtime benchmark (#6522)
NATURE committed
497c537
H
[Diffusion] Register HWR mmap for direct DLO H2D (#6591)
Hongsheng Liu committed
c608910
H
[Model] Optimize MiniMax-H3 DLO component lifecycle (#6526)
Hongsheng Liu committed
5a8b81e
N
[CI] Reduce Qwen-Image Function and share step-execution perf server (#6613)
NumberWan committed
dd527b4
J
[AutoRound] Add offline quantized MXFP4 model support (#5544)
jl9876 committed
0d5d111
P
[CI/Build] Clean up Qwen3-Omni nightly tests (#6570)
psv666 committed
683f3c9
J
[Misc] move text-to-audio online examples to a unified folder (#4807)
Jae committed
72beeae
B
[CI] Fix response_format json_schema error expectation (#6290)
Bhavyansh Sharma committed
d6b66fa
N
[BugFix] Qwen-image performance regressed - Avoid mapping diffusion_batch_size onto scheduler max_num_seqs (#6525)
NumberWan committed
bc8ea48
Z
[Test] add marker for stability and add Wan2.2 npu test case (#6343)
zhumingjue138 committed
29e3165
B
[perf] Add mask-free TRTLLM packed-padding path (#6542)
Bo Li committed
51b7565
W
[Refactor][2/N Scheduler]Replace _FULL_PAYLOAD_INPUT_STAGES with a resolved stage transport capability. (#6149)
wy17003 committed
b454e34
N
[minor, fix] Remove cuda sync on wake_up in AsyncOmni (#4092)
Nguyen Kha Nhat Long committed
bae58f8
W
[Bugfix] Fix remote replica membership lifecycle races (#5277)
Wei committed
f0808c9
N
[Feature][MiniCPM-o] Support configurable concurrent duplex sessions (#6021)
NATURE committed
702a409
Y
[Bugfix] Honor height/width overrides in LongCatImageEditPipeline (#6222)
yzong-rh committed
710024f
P
[Bugfix] Invalidate stale ref_audio cache on local file modification (#4873) (#5670)
Pranav Thakur committed
53b60bd