microsoft/onnxruntime

Pull Requests Commits

Adds ATen fallback for scaled_dot_product_attention (#21107)

prathikr committed 1 year ago

Verified 11ad2994

Fix typos according to reviewdog report. (#21335)

mindest committed 1 year ago

Verified 5b9369e9

Replace inline pip install with pip install from requirements*.txt (#21106)

jchen351 committed 1 year ago

Verified 4e75605e

[WebNN EP] Add outputDataType option for the ArgMax/ArgMin ops (#21385)

Honry committed 1 year ago

Verified 17e9ea62

[CUDA] FusedMHARunnerFP16v2 thread-safe (#21420)

tianleiwu committed 1 year ago

Verified a6c5e2cd

add transform part of the dq matmul tool chain (#21374)

fajin-corp committed 1 year ago

Verified 11bf3097

[TensorRT] Enable refitting an embedded engine when provided as byte stream (#21357)

gedoensmax committed 1 year ago

Verified 5bec5220

Add CoreML ML Program Resize (#21370)

skottmckay committed 1 year ago

Verified 34cd2e8e

[CUDA] Attention kernel provider option (#21344)

tianleiwu committed 1 year ago

Verified 6ffaaebb

[js/web] fix vulnerable version of dependencies (#21412)

fs-eire committed 1 year ago

Verified 01df8c78

Move ReluQuantFusion to Level2 for CPU EP only (#21329)

adrianlizarraga committed 1 year ago

Verified 22d4d82f

Enabled more VitisAI backend compilers (#21411)

glen-amd committed 1 year ago

Verified cc4049af

Update azure-kusto-data and azure-kusto-ingest (#21409)

Changming Sun committed 1 year ago

Verified 9140d9b1

[MLAS][AArch64] SQNBitGemm CompInt8 - Use 4x2 tiles (#21380)

edgchen1 committed 1 year ago

Verified 05fc0c60

remove llama 70b (#21396)

frank-dong-ms committed 1 year ago

Verified 92f66de7

[TensorRT EP] support TensorRT 10.2-GA (#21395)

yf711 committed 1 year ago

Verified bb76ead9

change ci docker image to rocm6.1 (#21296)

kailums committed 1 year ago

Verified 1b38c055

Enablement of onnxruntime for AIX and fixing issues related to big-endian platform. (#21133)

ranjitshs committed 1 year ago

Verified 6c7562b0

[ROCM] adjust test_flash_attn_rocm test tolerance (#21379)

tianleiwu committed 1 year ago

Verified 0f4c39ec

Add ML Program support for transpose op (#21364)

vraspar committed 1 year ago

Verified fa287042

Exclude blkq4_fp16_gemm_sm80_test in cuda 12.5 build (#21373)

tianleiwu committed 1 year ago

Verified 760a31c8

[VitisAI] fix graph save (#21293)

BoarQing committed 1 year ago

Verified dcc04367

matmul 4bit tool chain support qdq (#21362)

fajin-corp committed 1 year ago

Verified 5df4ddd1

Fix a build error when CUDA is enabled and onnxruntime_DISABLE_CONTRIB_OPS is ON (#21285)

Changming Sun committed 1 year ago

Verified 8568a676

Add ML Program support for basic activation ops (#21326)

vraspar committed 1 year ago

Verified 21830140

add vitisai ep build stage to Windows CPU Pipeline (#21361)

jywu-msft committed 1 year ago

Verified 4005d12e

Revert "Fix ETW Sink Initialize unproperly locking" (#21360)

adrianlizarraga committed 1 year ago

Verified cf565e95

[Optimizer] DQ + MatMul to MatMulNBits support: kernel changes (#21342)

fajin-corp committed 1 year ago

Verified 50170c69

Combining android build and test step into one job (#21340)

jchen351 committed 1 year ago

Verified c03e6fff

[TensorRT EP] c4996 suppression to build with trt10.2ga on Windows (#21358)

yf711 committed 1 year ago

Verified db9ee359

Newer Older