microsoft/onnxruntime

Pull Requests Commits

Fix required input validation CI

baijumeswani committed 48 days ago

83c0b7da

Merge branch 'main' of https://github.com/microsoft/onnxruntime into baijumeswani/fix-cuda-plugin-missing-input

baijumeswani committed 48 days ago

09fef2c2

Simplify required input collection

baijumeswani committed 48 days ago

3e4b84b2

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

bmehta001 committed 49 days ago

Verified 9aa8fb06

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

baijumeswani committed 49 days ago

Verified 2a9bf1e9

Support LoRA adapters with plugin EP allocators (#32221)

baijumeswani committed 49 days ago

Verified 4f7d0f5b

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

baijumeswani committed 49 days ago

Verified 15920d08

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

edgchen1 committed 49 days ago

Verified ddbcef7d

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

4n4ny4 committed 50 days ago

Verified 4d308dac

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

xiaofeihan1 committed 51 days ago

Verified 89b01b40

Fix C API docs Doxygen download (#32210)

Wayne-Ch committed 51 days ago

Verified df012c2f

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

mirounga committed 51 days ago

Verified a4156be6

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

mirounga committed 51 days ago

Verified 27f212c4

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

4n4ny4 committed 51 days ago

Verified 775526b8

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

4n4ny4 committed 51 days ago

Verified 8772f770

Pin GitHub Actions to full-length commit SHAs (#32176)

danfiedler-msft committed 51 days ago

Verified 5fd8e199

Resolve possible onnx CVE in python docs (#32190)

eserscor committed 51 days ago

Verified 1fbac8f3

[MLAS] AVX-512 16-wide Erf kernel and NCHWc reorder transpose for MobileClip-S0 model (#31958)

swetha097 committed 52 days ago

Verified a88c1168

[WebGPU] Optimized PagedAttention implementation (2/n) (#31727)

hariharans29 committed 52 days ago

Verified ebd7bad1

[MLAS] Hardswish Fusion implement for Mobilenetv3 models (#31957)

swetha097 committed 52 days ago

Verified 4db33198

Rework agent code review guidance (#31969)

edgchen1 committed 52 days ago

Verified 5d9676ea

Harden Crop operator validation (#32157)

edgchen1 committed 52 days ago

Verified 71f7c781

Validate per-element split sizes on the input-tensor path to prevent OOB read (#29461)

chilo-ms committed 52 days ago

Verified abc74dc0

Retain Python async run resources (#32041)

apsonawane committed 52 days ago

Verified 1cb4307d

Validate ScatterND index depth (#32034)

apsonawane committed 52 days ago

Verified d63396d9

Validate QEmbed segment inputs (#32144)

apsonawane committed 52 days ago

Verified 924b4d1d

Validate empty reduction axes (#32156)

apsonawane committed 52 days ago

Verified 113375ed

Validate Conv bias size (#32160)

apsonawane committed 52 days ago

Verified d9872d8e

Reject scalar Normalizer inputs (#32161)

apsonawane committed 52 days ago

Verified c6a81e4b

Skip overridable initializer fusion (#32143)

apsonawane committed 52 days ago

Verified 8805b399

Older