Commits microsoft/onnxruntime

Add agent review guidance for the native WebGPU EP

devang-ml committed 2 days ago

4453ec99

[WebGPU]: Add Intel subgroup prefill flash-attention kernel (#32702)

daijh committed 2 days ago

Verified ef3d2493

[WebGPU] Add enableFeatures and disableFeatures for EP feature management (#33481)

daijh committed 2 days ago

Verified e982a8bf

[CUDA] Accelerate external weight loading with pipelined pinned staging and plugin EP support (#33839)

tianleiwu committed 2 days ago

Verified ad2d14b5

Use local Python script for minimal-build test preparation (#33846)

eserscor committed 2 days ago

Verified 8c1f2936

Revert "Fix/gate cuda ci registry auth" (#33894)

tianleiwu committed 2 days ago

Verified 4177f554

Fix/gate cuda ci registry auth (#33376)

apsonawane committed 2 days ago

Verified 3605f37e

Add pybindings for Model Packaging (#33640)

preetha-intel committed 2 days ago

Verified cd18ed3c

Fix shape mismatch (#29863)

nazanin-beheshti committed 2 days ago

Verified ce2b0d3d

[WebGPU] Add int32 support to Relu opset 14 (#33789)

adrastogi committed 2 days ago

Verified ee9440b6

[WebGPU] Support int8 data in Gather (#33790)

adrastogi committed 2 days ago

Verified dd1a1659

[CPU/MLAS] Add QMoE shared layout INT2 AVX-512/AVX2 kernels and direct CPU QMoE (#33785)

tianleiwu committed 2 days ago

Verified f340e685

[CUDA] Document INT2 numerical limits and harden QMoE tests (#33858)

tianleiwu committed 2 days ago

Verified db4a8b1e

Bump joi from 17.13.7 to 17.13.9 in /js/react_native/e2e (#33797)

dependabot[bot] committed 3 days ago

Verified 7f16fabd

[WebGPU] Unroll bounded NVIDIA FlashAttention reductions in WGSL (#33395)

gyagp committed 3 days ago

Verified 1082ab76

[CUDA] Add INT2 QMoE CPU/GPU offload support and optimize spin-pool wakeup (#33787)

tianleiwu committed 3 days ago

Verified 3eb27a03

[CUDA] Fall back to FP32 in DSA INT8 TC kernel when rescale can flush a large-value-scale accumulator (#33847)

tianleiwu committed 3 days ago

Verified 1aa82412

[CUDA] Address review follow-ups from #33731 for QMoE offload and RouterGateUpProj (#33850)

tianleiwu committed 3 days ago

Verified f6acc099

Build WebGPU plugin EP separately for Windows documentation generation (#33829)

edgchen1 committed 3 days ago

Verified df4b8ac0

[CUDA] Enable Tensor-Core INT3/INT4 QMoE offload decode/prefill and 8K single-tile streaming for Qwen3.8 RC5 (#33731)

tianleiwu committed 3 days ago

Verified ec680674

Remove unused light-server from Web E2E dependencies (#33835)

eserscor committed 3 days ago

Verified c9c4e52b

Optimize CUDA float Pow with a guarded reciprocal square root fast path (#33662)

justinchuby committed 3 days ago

Verified a043a5fd

[CUDA] Optimize Qwen3.8-Flash-Next DSA, Indexer, HC, and QMoE kernels for SM12x and 256K context (#33802)

tianleiwu committed 3 days ago

Verified f7e70ad5

Isolate volatile build provenance from shared configuration (#33687)

justinchuby committed 3 days ago

Verified b2021624

Fix cmd escaping in Windows CMake install validation (#33840)

eserscor committed 3 days ago

Verified 07354eaf

Add vcpkg binary caching to Azure Pipelines (#33691)

eserscor committed 3 days ago

Verified 643d6ba7

[CUDA] Add BitwiseAnd, BitwiseOr, and BitwiseNot integer kernels (#33646)

justinchuby committed 3 days ago

Verified 242b1a91

Make misspell lint actionable: PR-only, blocking, no false positives (#33779)

adrastogi committed 3 days ago

Verified df5ef89b

Fix pooled GQA model buffer detachment in WebAssembly proxy tests (#33838)

eserscor committed 3 days ago

Verified f93e05cf

Handle kernel type string resolution errors without throwing (#33828)

mustjab committed 3 days ago

Verified 1aae736a