Skip to content

Pull requests: microsoft/onnxruntime-genai

New pull request New
26 Open 1,369 Closed
26 Open 1,369 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Route pipeline model RunOptions through SetRunOption for proper special key handling
#2044 opened Mar 24, 2026 by Copilot AI Loading…
@baijumeswani
2
Add Qwen3.5 hybrid decoder export support (GatedDeltaNet + Attention)
#2043 opened Mar 24, 2026 by apsonawane Loading…
7
Rename NemotronCacheConfig to NemotronConfig and add blank penalty to the decoder
#2042 opened Mar 22, 2026 by nenad1002 Loading…
3
Decouple plugin execution providers (EPs) from the USE_WINML pre-processor macro
#2038 opened Mar 19, 2026 by baijumeswani Loading…
62
Add Silero VAD Support to Nemotron Streaming ASR
#2035 opened Mar 17, 2026 by sayanshaw24 Loading…
49
Add WebGPU EP support and repetitions flag to whisper.py
#2032 opened Mar 17, 2026 by qjia7 Loading…
5
[VitisAI] external_ep_library typo fix
#2027 opened Mar 13, 2026 by akholodnamdcom • Draft
1
Add Qwen3.5 support
#2025 opened Mar 13, 2026 by kinfey Loading…
9
[WebGPU] Add compact attention mask support
#2022 opened Mar 13, 2026 by qjia7 Loading…
8
Support Visual Studio 18 2026 build
#2017 opened Mar 11, 2026 by Copilot AI • Draft
@xiaofeihan1
2
Add support for QNN stateful models
#2012 opened Mar 6, 2026 by qti-ashimaj Loading…
17
[Don't review] Optimizations for graph capture
#2011 opened Mar 6, 2026 by qjia7 • Draft
Handle VLM weight name prefixes in QuantizedModel loader
#1996 opened Feb 28, 2026 by uday610 • Draft
GenAI changes to support EPContext compilation and validation
#1993 opened Feb 27, 2026 by lnigam Loading…
29
Add model builder support for LFM2
#1979 opened Feb 14, 2026 by xenova Loading…
21
[Draft] Parakeet export
#1977 opened Feb 12, 2026 by jiafatom • Draft
10
remove one assert not verified with model microsoft/OptiMind-SFT
#1975 opened Feb 12, 2026 by xadupre Loading…
KV Cache optimization based on sequence length
#1974 opened Feb 11, 2026 by chilukam-qti Loading…
Support UpdatePositionIds in webgpu's interface
#1952 opened Jan 22, 2026 by qjia7 • Draft
3
Make phi model of webgpu ep always use long RoPE to improve tps performance for long context scenario
#1930 opened Dec 22, 2025 by feich-ms • Draft
3
Remove unused unit tests
#1923 opened Dec 17, 2025 by apsonawane Loading…
1
Also clear provider_options when ClearProviders is called
#1894 opened Nov 26, 2025 by Zhaeong Loading…
Enable GENAI with FetchContent or find_package() for client projects
#1858 opened Nov 9, 2025 by apwojcik Loading…
Support for decoder pipeline based multimodel
#1775 opened Sep 18, 2025 by prudhvi-qti • Draft
9
Modify Model Builder to build paged attention models
#1605 opened Jul 3, 2025 by aciddelgado • Draft
9
ProTip! Follow long discussions with comments:>50.