All pull requests

Pull requests

• Retain optimized models when quantization shape inference is skipped

  #32803·Sylvester Kaczmarek (sylvesterkaczmarek) opened Sep 25, 2026Contributor·

  ·
  1

  5

• [CUDA] Disable unsafe graph replay in plugin EP

  #32801·Tianlei Wu (tianleiwu) opened Sep 25, 2026Contributor·

  ·
  2

• Reduce coordinate work in CPU N-D Col2Im

  #32800· Stern (Hosi121) opened Sep 25, 2026Contributor·

  ·
  2

• test fix implemented

  #32798·Melike Kaptan (melkap01-Arm) opened Sep 25, 2026Contributor·

  ·
  2

• Only request timestamp-query in WebGPU EP if profiling is enabled for the session

  #32797·Scott McKay (skottmckay) opened Sep 25, 2026Contributor·

  ·
  7

• [WebGPU] Optimize subgroup matrix MatMulNBits for AMD wave64

  #32796·Yang Gu (gyagp) opened Sep 25, 2026Contributor·

  ·
  4

• [WebGPU] Package D3D12 Agility SDK for Python wheels

  #32795·May Mei (mei1127) opened Sep 25, 2026Contributor·

  ·
  10

• Add WebGPU block-scaled FP8 QMoE

  #32794·Sayan Shaw (sayanshaw24) opened Sep 24, 2026Contributor·

  ·
  7

• Bump image-size, react-native and @react-native/metro-config in /js/react_native/e2e

  dependenciesjavascript
  #32793·dependabot[bot] opened Sep 24, 2026Bot·

  ·
  3

• Local non-causal for paged attention

  #32792·aciddelgado opened Sep 24, 2026Contributor·

  ·
  3

• Add github action for arm64 windows build

  #32791·eserscor opened Sep 24, 2026Contributor·

  ·
  1

• Fix model package EP setup and experimental API contracts

  #32790·Xiaoyu Z (xiaoyu-work) opened Sep 24, 2026Member·

  ·

• [WebGPU] Allow exact runtime compatibility exceptions

  #32787·Xiaofei Han (xiaofeihan1) opened Sep 24, 2026Contributor·

  ·
  2

• [WIP] Fix CPUExecutionProvider output differences for FusedConv model

  #32782·tianleiwu with CopilotAI opened Sep 24, 2026·

  ·
  1

  1

  

• Default Java GPU packaging to CUDA 13

  #32781·tianleiwu with CopilotAI opened Sep 24, 2026·

  ·
  1

  1

  

• Preserve EP registration End events for returned errors without telemetry

  #32780·bmehta001 opened Sep 24, 2026Contributor·

  ·
  1

  

• Add ability for app to pin ORT version when consuming onnxruntime-react-native npm package

  #32779·Scott McKay (skottmckay) opened Sep 24, 2026Contributor·

  ·
  1

• [Plan] Multi-device EP support

  #32778·Scott McKay (skottmckay) opened Sep 24, 2026Contributor·

  ·

• fix: name the activation the WebGPU Conv kernel cannot fuse (#32716)

  #32775·Yuri Khrustalev (ykhrustalev) opened Sep 24, 2026Contributor·

  ·
  2

• Fold Transpose on constant initializer instead of pushing it

  #32770·Yuduo Wu (1duo) opened Sep 23, 2026Contributor·

  ·
  8

• Test Windows 2025 migration

  #32769·eserscor opened Sep 23, 2026Contributor·

  ·

• Split webgpu build into build and test. Reduce total runtime 30~45 mins.

  #32768·eserscor opened Sep 23, 2026Contributor·

  ·
  8

• [WebGPU EP] adds state update capacity support for GatedDeltaNet

  #32767·Prathik Rao (prathikr) opened Sep 23, 2026Contributor·

  ·
  13

• Enable non-causal WebGPU PagedAttention for full-context blocks

  #32763·aciddelgado opened Sep 23, 2026Contributor·

  ·
  7

• [WIP] Fix failing GitHub Actions job Build Linux x64 Release

  #32762·mirounga with CopilotAI opened Sep 23, 2026·

  ·
  1