Pull Requests microsoft/onnxruntime

Fix CPU LSTM input validation and refactor ValidateInputs to frontends

#28653 opened 2026-05-23 14:19 by vacu9708

fix 302 redirects while postinstall, with help of axios

#28652 opened 2026-05-23 14:14 by phantom9999

Fix: add missing fields to ToProviderOptions for CUDA

#28651 opened 2026-05-23 09:28 by G3nna

js/node: offload Run() to thread pool via napi_async_work, add RunSync

#28646 opened 2026-05-22 21:35 by A-KGeorge

Add multi-shape profiling support to perftest tool (`--data_shape`)

#28638 opened 2026-05-22 08:35 by adrianlizarraga

Bump uuid, detox and jest-junit in /js/react_native/e2e dependencies javascript

#28616 opened 2026-05-21 20:13 by dependabot[bot]

[CUDA Plugin EP] Add per-node attribution and explicit GPU→ORT event linkage to profiler

#28614 opened 2026-05-21 19:27 by tianleiwu

[WebGPU] Avoid indirect dispatch in FlashAttention decode to fix perf issues with Vulkan backend + GraphCapture/GraphReplay ep:WebGPU

#28581 opened 2026-05-20 06:45 by hariharans29

TurboQuant KV cache (4/4): Python reference impl + last_token_logits patcher

#28563 opened 2026-05-19 11:32 by TimPietrusky

TurboQuant KV cache (3/4): WebGPU kernels + Safari/Firefox fallback ep:WebGPU

#28562 opened 2026-05-19 11:31 by TimPietrusky

TurboQuant KV cache (2/4): CUDA kernels

#28561 opened 2026-05-19 11:29 by TimPietrusky

TurboQuant KV cache (1/4): graph rewrite + schema (foundation)

#28560 opened 2026-05-19 11:28 by TimPietrusky

[docs] Add WebGPU EP docs

#28553 opened 2026-05-19 01:56 by edgchen1

Reserve WebGPU program vector capacity

#28534 opened 2026-05-17 20:52 by Lingavasan

fix(cann): prevent race condition on .om cache file during multi-card compilation

#28533 opened 2026-05-17 14:16 by 0AnshuAditya0

Fix MLAS LUT GEMM to support 2-bit MatMulNBits with N not divisible by 128

#28528 opened 2026-05-15 23:14 by vraspar

Add per-I/O memtype controls for plugin EPs

#28514 opened 2026-05-14 18:38 by ericcraw

[webgpu] Add NTC layout support for CausalConvWithState ep:WebGPU

#28504 opened 2026-05-14 07:35 by xiaofeihan1

Implement staged CI pipeline: Linux CPU → Linux CUDA → everything else

#28493 opened 2026-05-13 19:11 by Copilot

[MLAS][KleidiAI] Add Arm64 fp16 MLAS and KleidiAI support

#28487 opened 2026-05-13 09:06 by Laan33

Honor Reshape allowzero in symbolic_shape_infer

#28470 opened 2026-05-12 13:41 by 1fanwang

Add C++ in-code profiling example code to the "Profiling Tools" documentation page

#28464 opened 2026-05-11 21:29 by kareertl

Add set_default_logger_callback Python API for custom ORT logging

#28457 opened 2026-05-11 17:25 by Copilot

Add transformer optimizer JSON report

#28445 opened 2026-05-10 03:10 by KOKOSde

fix: use relocatable path for safeint_interface when using vcpkg

#28444 opened 2026-05-09 18:46 by 2ssh

Fix f16 overflow in SkipLayerNormalization CUDA kernel

#28442 opened 2026-05-09 16:24 by justinchuby

Fix CPU LayerNormalization NaNs on large constant inputs

#28440 opened 2026-05-09 15:59 by Copilot

webgpu: support vec2 type for gemm-subgroup ep:WebGPU

#28436 opened 2026-05-09 07:49 by xhcao

[MLAS] Add CPU DynamicQuantMatMulFp8 contrib op with MLAS FP8 fallback

#28416 opened 2026-05-08 15:37 by melkap01-Arm

Bump diffusers from 0.28.0 to 0.38.0 in /onnxruntime/python/tools/transformers/models/stable_diffusion/requirements dependencies python

#28401 opened 2026-05-07 18:33 by dependabot[bot]