Morph MORPH
SIGN IN SIGN UP
dhruvhead / onnxruntime UNCLAIMED
0 0 90 C++

COMMITS

main
May 19, 2026
A
Add EP and hardware device type to Windows ML telemetry (#28477)
angelser committed
4d1dce8
Q
[MLAS] test: accept canonical NaN in activation NaN round-trip check (#28538)
qiurui144 committed
b38731b
A
Validate sparse tensor external file paths (#28408)
Adrian Lizarraga committed
6a8f021
D
Fix XNNPACK Gemm SIGSEGV on missing and scalar 'C' bias (#28546)
Dhruvil Parikh committed
8c650d8
C
Fix int32 overflow in CUDA Cast and UnaryElementWise kernels for tensors with >2^31 elements (#28386)
Copilot committed
cd05ef4
T
[Security] Validate CPU cache_indirection beam indices (#28486)
Tianlei Wu committed
2d9f35f
S
Skip file path validation in CreateEpContextModel when the output is not a file (#28431)
shiyi committed
ce6aaee
A
Add unit test for fix of plugin EP library handle leak in PR 28396 (#28430)
Adrian Lizarraga committed
40e3048
May 18, 2026
D
Support kquant mixed for whisper export (#28515)
David Fan committed
d2453bc
C
contrib: add mask/input shape consistency checks in MaxpoolWithMask::Compute (#28223)
Copilot committed
9883e93
T
Reject recursive local function definitions during model load (#28187)
Tianlei Wu committed
22e537b
E
[WebGPU plugin EP] Package improvements (#28525)
Edward Chen committed
770da7d
J
webgpu: Generalize FlashAttention prefill shared-memory path (#28520)
Jiajia Qin committed
aa416f5
L
fix: avoid vector copies in CheckIfSubtreesAreEqual (#27854)
Luis Rios committed
4f620c3
May 15, 2026
C
Fix Inverse kernel rank underflow before indexing trailing dims (#28400)
Copilot committed
b429dff
H
[MLAS] Fix scratch buffer over-allocation in batched/grouped NCHW Conv (#28482)
Hariharan Seshadri committed
eb63cc6
R
Fix mlasi_sve.h preprocessor guards to allow clang compilation (#28507)
Ryan VanderMeulen committed
45c0663
A
[QNN EP] ETW log level rule change (#27593)
Ashwath Shankarnarayan committed
ad6d34c
V
Fix input validation and null-pointer dereference in STFTDecomposition graph transformer (#28465)
vraspar committed
00e9ef3
B
Fix WeaklyCanonicalPath ERROR_ACCESS_DENIED in Windows AppContainers (#28509)
Brenden Sosnader committed
ddd1eae
K
Build issues on AIX for POWER10 and POWER11 (#26704)
KV2773 committed
127faf0
May 14, 2026
C
Fill LSTM CUDA operator opset gap: extend coverage from opset 14 to opset 22 (#27737)
Copilot committed
c0b3212
E
Validate per-column weight_scale/weight_zero_point shape in CPU QAttention; harden integer arithmetic in QAttention and AttentionBase (#28480)
Edward Chen committed
18ae6da
G
enable dynamic max_k_step in FA for nvidia (#28511)
Guenther Schmuelling committed
b36d8db
O
[MLAS] Add an NHWC implementation of convolution to avoid transposes (#26834)
Orlaith Monahan committed
a8ba94a
D
Improve error reporting for pre-allocated outputs with a wrong shape (#28481)
Dmitri Smirnov committed
70dcc0e
T
fix(security): add SafeInt overflow protection in Expand and constant folding output size limit (#28055)
Tianlei Wu committed
04f7440
R
fix: support N-D weights with unit leading dims in MatMulNBitsQuantizer (#28178)
Rishi Dave committed
2979bab
C
Fix out-of-bounds read in CropBase: validate scale attribute length (#28399)
Copilot committed
50d04b9
T
CUDA Plugin Packaging Pipeline Fix TSAUpload and Timeout (#28497)
Tianlei Wu committed
dcf8775