Search results

[Feature Request] Support BF16 for ConvTranspose-22 in CUDA

ep:CUDAissues related to the CUDA execution providerissues related to the CUDA execution provider

Status: Open.

#28476 In microsoft/onnxruntime;

· tianleiwu opened on May 12, 2026

Cara Buka Mobile Token BNI Direct terkunci

documentationimprovements or additions to documentation; typically submitted using templateimprovements or additions to documentation; typically submitted using template

platform:mobileissues related to ONNX Runtime mobile; typically submitted using templateissues related to ONNX Runtime mobile; typically submitted using template

Status: Open.

#28473 In microsoft/onnxruntime;

· callcenter559 opened on May 12, 2026

[Build] Windows release lacks cmake files

buildbuild issues; typically submitted using templatebuild issues; typically submitted using template

platform:windowsissues related to the Windows platformissues related to the Windows platform

Status: Open.

#28468 In microsoft/onnxruntime;

· g-gisbert opened on May 12, 2026

[Performance] Speed-vs-memory regression on Conv3D FusedConv as of version 1.23.2

performanceissues related to performance regressionsissues related to performance regressions

Status: Open.

#28466 In microsoft/onnxruntime;

· ryanseghers opened on May 11, 2026

LayerNormalization returns NaN for finite large float32 inputs

Status: Open.

#28463 In microsoft/onnxruntime;

· ALinrunrun opened on May 11, 2026

Erf returns 1.0 instead of NaN for NaN inputs

Status: Open.

#28462 In microsoft/onnxruntime;

· ALinrunrun opened on May 11, 2026

ReduceLogSumExp returns NaN for all -inf input

Status: Open.

#28461 In microsoft/onnxruntime;

· ALinrunrun opened on May 11, 2026

Large numerical discrepancy in CPUExecutionProvider ReduceSum for large float32 tensor

Status: Open.

#28450 In microsoft/onnxruntime;

· tinywisdom opened on May 11, 2026

Reshape with allowzero=1 produces wrong shape for zero-size tensor

Status: Open.

#28449 In microsoft/onnxruntime;

· ALinrunrun opened on May 10, 2026

ORT_ENABLE_ALL produces incorrect output for QuantizeLinear-DequantizeLinear Resize(linear) pattern

quantizationissues related to quantizationissues related to quantization

Status: Open.

#28448 In microsoft/onnxruntime;

· ALinrunrun opened on May 10, 2026

CPU QLinearConv rejects per-channel weight zero points with different values

Status: Open.

#28447 In microsoft/onnxruntime;

· ALinrunrun opened on May 10, 2026

Can't run htdemucs with MIGraphX

ep:MIGraphXissues related to AMD MI GraphX execution providerissues related to AMD MI GraphX execution provider

Status: Open.

#28446 In microsoft/onnxruntime;

· jclsn opened on May 10, 2026