Skip to content

Pull requests: microsoft/onnxruntime

New pull request New
461 Open 18,533 Closed
461 Open 18,533 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

QMoE CUDA: input validation, prepack cleanups, and packaging pipeline fix
#28607 opened May 21, 2026 by tianleiwu Contributor Loading…
4 tasks done
Optimize MLAS quantized KV-cache GEMM kernels (follow-up to #28578)
#28606 opened May 21, 2026 by tianleiwu Contributor Loading…
[OpenVINO EP] Validate that EPContext binary path is within model con…
#28605 opened May 21, 2026 by adrianlizarraga Contributor • Draft
10
[CUDA Plugin EP] Add provider options: user_compute_stream, do_copy_in_default_stream, use_ep_level_unified_stream, external allocator
#28603 opened May 20, 2026 by tianleiwu Contributor Loading…
3 tasks done
5
Move OrtEp sanity checks to plugin EP creation and remove check for OrtEp::ort_version_supported upper bound
#28601 opened May 20, 2026 by edgchen1 Contributor Loading…
1
Fix Whisper genai_config context_length using wrong config attribute
#28600 opened May 20, 2026 by jiafatom Contributor Loading…
Add component governance manifest for WebGPU EP
#28599 opened May 20, 2026 by adrastogi Contributor Loading…
21
[CoreML EP] Add GatherND builder
#28598 opened May 20, 2026 by maxwbuckley Contributor • Draft
[CoreML EP] Add Where and And builders
#28597 opened May 20, 2026 by maxwbuckley Contributor • Draft
[CoreML EP] Add Sin and Cos unary ops
#28596 opened May 20, 2026 by maxwbuckley Contributor • Draft
[CoreML EP] Support bool Cast in ML Program
#28595 opened May 20, 2026 by maxwbuckley Contributor • Draft
Optimize MatMulNBits 2-bit + float zero_point CPU dequantization with multi-threaded kernel
#28589 opened May 20, 2026 by Copilot AI Loading…
1
@tianleiwu
12
Parallelize CPU ScatterElements kernel via ThreadPool
#28588 opened May 20, 2026 by Copilot AI Loading…
1
@tianleiwu
8
CPU ReduceSum: improve large float32 full-reduction accuracy and add regression coverage
#28587 opened May 20, 2026 by Copilot AI • Draft
1
@justinchuby
2
Build with TensorRT 11 and abseil 20250814 (NVCC)
#28586 opened May 20, 2026 by mc-nv Contributor Loading…
4
[WebGPU] Avoid indirect dispatch in FlashAttention decode to fix perf issues with Vulkan backend + GraphCapture/GraphReplay ep:WebGPU ort-web webgpu provider
#28581 opened May 20, 2026 by hariharans29 Member Loading…
3
Bump protobufjs from 7.2.5 to 7.5.8 in /js/web dependencies Pull requests that update a dependency file javascript Pull requests that update Javascript code
#28573 opened May 19, 2026 by dependabot Bot Loading…
[MLAS] KleidiAI fix igemm regression
#28571 opened May 19, 2026 by martin-klacer-arm Loading…
1
7
[MLAS] Update the NHWC sans transposes path to also support Depthwise convolutions
#28565 opened May 19, 2026 by orlmon01 Contributor Loading…
5
TurboQuant KV cache (4/4): Python reference impl + last_token_logits patcher
#28563 opened May 19, 2026 by TimPietrusky • Draft
1
TurboQuant KV cache (3/4): WebGPU kernels + Safari/Firefox fallback
#28562 opened May 19, 2026 by TimPietrusky • Draft
1
TurboQuant KV cache (2/4): CUDA kernels
#28561 opened May 19, 2026 by TimPietrusky • Draft
1
TurboQuant KV cache (1/4): graph rewrite + schema (foundation)
#28560 opened May 19, 2026 by TimPietrusky • Draft
1
Raise protobuf minimum version in Python requirements
#28558 opened May 19, 2026 by anzzraju1997-glitch Loading…
1
Fix CUDA build with contrib ops disabled
#28554 opened May 19, 2026 by Copilot AI Loading…
1
@tianleiwu
9
ProTip! Add no:assignee to see everything that’s not assigned.