microsoft/onnxruntime

Pull Requests Commits

Update C# callbacks for opaque read options

Gopalakrishnan Nallasamy committed 7 hours ago

e9732c8b

Add C# EPContext data callbacks

Gopalakrishnan Nallasamy committed 9 hours ago

576b6386

Rerun infrastructure-failed CI

Gopalakrishnan Nallasamy committed 9 hours ago

442ca91b

Fix EPContext tests in minimal and GPU builds

Gopalakrishnan Nallasamy committed 1 day ago

bbdcdd83

Address EPContext API review feedback

Gopalakrishnan Nallasamy committed 1 day ago

af34dcb5

Guard EPContext test size constant

Gopalakrishnan Nallasamy committed 1 day ago

1e0032dd

Fix EPContext self-move test warnings

Gopalakrishnan Nallasamy committed 1 day ago

143ea52b

Address EPContext stable API review feedback

Gopalakrishnan Nallasamy committed 2 days ago

d5a04e50

Fix EPContext callbacks in minimal builds

Gopalakrishnan Nallasamy committed 2 days ago

207f00d7

Promote EPContext data callbacks to stable API

Gopalakrishnan Nallasamy committed 2 days ago

4576aba5

Run wgsl_template Python tests in CI (#32214)

edgchen1 committed 2 days ago

Verified dc999734

[MLAS] Reject Arm® KleidiAI™ Q4 prepack with dynamic scales (#32068)

martin-klacer-arm committed 3 days ago

Verified 9cbf665b

[MLAS] Refactor QNBit KleidiAI integration into dedicated layer (#31720)

martin-klacer-arm committed 3 days ago

Verified 5cb241b5

Validate MatMulNBits 8-bit g_idx bounds on CUDA (#31643)

apsonawane committed 3 days ago

Verified 066b90ef

Fix Windows CUDA 12.9 SM120 compilation (#32114)

Copilot committed 3 days ago

Verified 939e180a

pin review dog (#32208)

eserscor committed 3 days ago

Verified ddc8f0c3

Upgrade Protobuf to v33.6 (#29906)

Copilot committed 3 days ago

Verified 993cd135

Extend memory importer with host pointer support (#29726)

javier-intel committed 3 days ago

Verified 710f0ce3

[Telemetry] Prevent crashes in shell-less containers (#32226)

bmehta001 committed 3 days ago

Verified 9aa8fb06

[CUDA] Add VarlenCausalConvWithState for continuous batching (#32168)

baijumeswani committed 3 days ago

Verified 2a9bf1e9

Support LoRA adapters with plugin EP allocators (#32221)

baijumeswani committed 3 days ago

Verified 4f7d0f5b

[CUDA] Extend paged XQA decode to head size 256 (#32229)

baijumeswani committed 3 days ago

Verified 15920d08

Move ort-release-notes skill to new .github/skills location. (#32193)

edgchen1 committed 3 days ago

Verified ddbcef7d

[WebGPU] Supply fused activation parameters to Conv/MatMul as uniforms (#32116)

4n4ny4 committed 5 days ago

Verified 4d308dac

[WebGPU] Scale Dawn pipeline compilation workers with CPU count (#29820)

xiaofeihan1 committed 6 days ago

Verified 89b01b40

Fix C API docs Doxygen download (#32210)

Wayne-Ch committed 6 days ago

Verified df012c2f

[ARM] MLAS: SVE i8mm (svmmla) int8 QGEMM kernels, portable machine code (#31146)

mirounga committed 6 days ago

Verified a4156be6

[MLAS] Add a NEON fused kernel for LinearAttention (#32178)

mirounga committed 6 days ago

Verified 27f212c4

[WebGPU] Regenerate WGSL in-tree golden fixtures to match current templates (#32115)

4n4ny4 committed 6 days ago

Verified 775526b8

[WebGPU] Make MatMulNaiveProgram's pipeline cache key cover everything it bakes into WGSL (#32048)

4n4ny4 committed 6 days ago

Verified 8772f770

Older