microsoft/onnxruntime

Pull Requests Commits

[webgpu] optimize SkipLayerNormalization operator (#24164)

xhcao committed 270 days ago

Verified 0acb0488

[webgpu] fix bias-add (#24336)

jchen10 committed 270 days ago

Verified 34abb8b4

[webgpu] Fix bias_split_gelu (#24342)

jchen10 committed 270 days ago

Verified 22656134

Remove explicit batch network flag for TRT 10+ (#24298)

kevinch-nv committed 271 days ago

Verified 4edada60

[webgpu] fix the reflect mode issue of Pad (#24202)

xhcao committed 271 days ago

Verified 10e51d26

webgpu support for DequantizeLinear (#24268)

guschmue committed 271 days ago

Verified f83e6618

Use WASM f32x4 relaxed min/max for relaxed simd build (#24324)

fanchenkong1 committed 271 days ago

Verified 04e0b50c

[webgpu] Flash attention for generation (#23808)

qjia7 committed 271 days ago

Verified 18f91e55

[VitisAI EP] export InferShapes to VitisAIEP (#23881)

zz002 committed 271 days ago

Verified 554fb4ad

[Native WebGPU] Exclude WebGPU EP from ConvFp16 3D tests. (#24327)

satyajandhyala committed 271 days ago

Verified b803429a

[webgpu] Fix ROUND_PREFER_CEIL issue of Resize operator (#24229)

xhcao committed 271 days ago

Verified ca1b32df

Implement load cancellation ability (#24257)

yuslepukhin committed 272 days ago

Verified 350d1400

[webgpu][dawn API optimization] reduce number of calls to buffer APIs (#24315)

fs-eire committed 272 days ago

Verified 73676fc5

[webgpu] Use 1D dispatch groups for attention (#24228)

qjia7 committed 272 days ago

Verified a1186f63

Add ConvTranspose cache key (#24317)

satyajandhyala committed 274 days ago

Verified d6df4f29

Fix 'minimal_power' to 'minimum_power' for DirectML performance selection (perf test) (#24303)

virajwad committed 274 days ago

Verified 55c1a3b0

[webgpu][dawn API optimization] reduce number of calls to wgpuDeviceGetQueue (#24313)

fs-eire committed 274 days ago

Verified a7e62d63

[Native WebGPU] Add Conv, ConTranspose and FusedConv (#24186)

satyajandhyala committed 274 days ago

Verified 9102aaee

Add support for uint8_t as data type for GatherBlockQuantized (#24239)

sushraja-msft committed 274 days ago

Verified a4976e33

Update packaging pipeline for Nodejs binding (#24301)

fs-eire committed 274 days ago

Verified 11fda2ad

Support Gemma3 with Clip fused attention (#24280)

titaiwangms committed 274 days ago

Verified e944379e

[WebGPU] fix cache key of AttentionProbs/VxAttentionScore (#24309)

fs-eire committed 274 days ago

Verified 2e94c5a4

Bump vite from 6.2.4 to 6.2.5 in /js/web/test/e2e/exports/testcases/vite-default (#24312)

dependabot[bot] committed 274 days ago

Verified 56f10183

[WebGPU] fix Pad cache key (#24305)

fs-eire committed 274 days ago

Verified 318cc87f

[QNN-EP] Fix ONNX context model helper. (#24271)

minfhong-quic committed 275 days ago

Verified 1cb53d00

upgrade action shellcheck to v1.30.0 (#24304)

fs-eire committed 275 days ago

Verified 82c8e569

[webgpu] Optimize MatMulNBits for f16 Block32 prefill performance (#23908)

daijh committed 275 days ago

Verified 3dfc2ae3

Update publish-nuget.yml to correct feed. (#24299)

jchen351 committed 275 days ago

Verified e5e906ee

[QNN-EP] Enhance QNN-EP support for Softmax with opset < 13. (#24180)

zhaoxul-qti committed 275 days ago

Verified a28da4b9

Bump image-size from 1.1.1 to 1.2.1 in /js/react_native/e2e (#24278)

dependabot[bot] committed 275 days ago

Verified cbaa8bc7

Newer Older