microsoft/onnxruntime

Pull Requests Commits

Enable Qnn nuget nightly (#20662)

HectorSVC committed 1 year ago

Verified 0e11d0c4

Nuget Publish pipelines should be trigger by rel-* automatically too. (#20652)

mszhanyi committed 1 year ago

Verified c131ea89

[js/webgpu] Support GroupQueryAttention (#20237)

axinging committed 1 year ago

Verified 8c59cd4f

Allow path pattern to be specified in package_release_tasks.py. (#20650)

edgchen1 committed 1 year ago

Verified 90d49ccb

Quant tool: make removal of Clip/Relu ops configurable (#20616)

adrianlizarraga committed 1 year ago

Verified 643ed147

Enable ClipQuantFusion exclusively on CPU EP (#20627)

yihonglyu committed 1 year ago

Verified 49d197a8

Java CUDA 12 support (#20583)

jchen351 committed 1 year ago

Verified 4fe565a6

[CUDA] Benchmark GQA on popular LLM models (#20646)

tianleiwu committed 1 year ago

Verified 85facd67

Generalize label input sparsity check and refactor (#20636)

guyang3532 committed 1 year ago

Verified cfe830b2

set unload to false to prevent crash when linux lib load not successfully (#20626)

smile-gator committed 1 year ago

Verified e124cf8e

Update SparseAttention op spec to make it more flexible (#20625)

tianleiwu committed 1 year ago

Verified 01dd991f

[qnn ep] sign onnxruntime.dll/pyd for qnn packages (#20634)

jywu-msft committed 1 year ago

Verified a0c4bd4d

Improve perf for mem efficient grad mgmt (#20480)

pengwa committed 1 year ago

Verified 56f70355

Migrate training storage from SAS to managed identity (#20618)

mszhanyi committed 1 year ago

Verified 5a18818e

Enable QNN HTP support for Node (#20576)

Jon Campbell committed 1 year ago

Verified 768c7931

The time for nuget pkg should be consistent (#20522)

jchen351 committed 1 year ago

Verified d1cbb3e0

[CUDA] Sparse Attention support 128k sequence length (#20614)

tianleiwu committed 1 year ago

Verified 69cfcba3

Update CocoaPods package release script. (#20608)

edgchen1 committed 1 year ago

Verified a0db2187

Fix SparseAttention cos/sin cache dimension checks (#20609)

kunal-vaishnavi committed 1 year ago

Verified 274d162d

support --arm64ec for qnn ep build (#20607)

jywu-msft committed 1 year ago

Verified 58d7b122

Implement numpy array over CPU OrtValues on return values (#20539)

yuslepukhin committed 1 year ago

Verified 08ecf30e

optimize gqa cpu (#20598)

yufenglee committed 1 year ago

Verified 156d5216

Fix GroupQueryAttention benchmark script (#20291)

tianleiwu committed 1 year ago

Verified 1f509215

Revert compute queue default for DML (#20604)

maggie1059 committed 1 year ago

Verified b6d9abf1

Remove an extra space in azure_scale_set_vm_mount_test_data.sh (#20584)

Changming Sun committed 1 year ago

Verified 08b63735

optimize skiplayernorm (#20551)

guschmue committed 1 year ago

Verified 55a6986d

MIGraphX EP: Add set_false_math to false by default (#20520)

TedThemistokleous committed 1 year ago

Verified 737eb48f

Clarify when protobuf dependency builds protoc (#20542)

skottmckay committed 1 year ago

Verified 8d09baf4

fix gqa cpu nan bug (#20521)

aciddelgado committed 1 year ago

Verified 4e27841b

Dev/mookerem/arm64x update (#20536)

moyo1997 committed 1 year ago

Verified aff04ba0

Newer Older