5 Pull requests merged by 5 people

• [WebNN] Support SkipSimplifiedLayerNormalization op

  #23151 merged Dec 24, 2024

• Integrate onnx 1.17.0

  #21897 merged Dec 24, 2024

• [DOC] update coreml new opretors and flags

  #22958 merged Dec 24, 2024

• [QNN EP] Fix multithread sync bug in ETW callback

  #23156 merged Dec 23, 2024

• Enable pointer-generator T5 models in BeamSearch

  #23134 merged Dec 23, 2024

10 Pull requests opened by 9 people

• Updated the Documentation for nuget packages

  #23182 opened Dec 23, 2024

• delete the supported domain version upper bounds

  #23184 opened Dec 24, 2024

• Fix the bug of C++ sample in CANN EP file

  #23186 opened Dec 24, 2024

• [webgpu] Use override shape in shader key

  #23188 opened Dec 24, 2024

• fix: Pad/AveragePool fusion

  #23190 opened Dec 24, 2024

• [WebGPU] Support PIX Capture for WebGPU EP

  #23192 opened Dec 25, 2024

• [js/webgpu] validate transpose perm if specified

  #23197 opened Dec 26, 2024

• [webgpu] Implement Split operator

  #23198 opened Dec 26, 2024

• [VitisAI] change all support tensor type from ir 9 to ir 10

  #23204 opened Dec 27, 2024

• [TensorRT EP] Fix to build ORT on legacy TRT8.5

  #23215 opened Dec 27, 2024

6 Issues closed by 4 people

• [Feature Request] Integrate with ONNX 1.17.0 release branch

  #22083 closed Dec 24, 2024

• [Build] Why calling the c++ interface only uses the CPU for inference

  #23159 closed Dec 24, 2024

• Error merging shape info for output

  #23185 closed Dec 24, 2024

• Support pointer-generator in BeamSearch op

  #23123 closed Dec 23, 2024

• ort-nightly venv install regressed

  #22922 closed Dec 23, 2024

• T5-Small different output for decoder inference with CPU and DirectML EPs

  #22896 closed Dec 21, 2024

24 Issues opened by 14 people

• [Feature Request] Inquiry About ONNX Runtime Support for Dynamic Decoding Correction in Machine Translation on Android

  #23216 opened Dec 28, 2024

• onnxruntime-web dependency on document breaks chrome serviceworker

  #23214 opened Dec 27, 2024

• RUNTIME_EXCEPTION : Non-zero status code returned while running If node.

  #23213 opened Dec 27, 2024

• ONNXRuntime produces inconsistent results for specific output v10_0 (flaky test behavior)

  #23212 opened Dec 27, 2024

• Inconsistent results with different optimization settings

  #23211 opened Dec 27, 2024

• ONNXRuntime Optimization Causes Output Discrepancy with Certain opt_level Settings

  #23210 opened Dec 27, 2024

• ONNXRuntime Optimization Causes Output Discrepancy in Specific Model Structure (Output Y)

  #23209 opened Dec 27, 2024

• Java library throws error when using CUDA: LoadLibrary failed with error 126 "" when trying to load "C:\Users\xx\AppData\Local\Temp\onnxruntime-java5278075328315693241\onnxruntime_providers_cuda.dll"

  #23208 opened Dec 27, 2024

• ONNXRuntime Optimization Causes Output Discrepancy in BiasDropout Operator

  #23207 opened Dec 27, 2024

• [Mobile] google say not support nnapi anymore

  #23206 opened Dec 27, 2024

• custom op's SUPPORTED_TENSOR_TYPES does not include int4 and uint4

  #23205 opened Dec 27, 2024

• [Inference Error] The onnx inference result is inconsistent with the numpy inference result

  #23202 opened Dec 26, 2024

• Different results between GPU and CPU

  #23201 opened Dec 26, 2024

• Inconsistent Results After ONNX Runtime Optimization

  #23200 opened Dec 26, 2024

• Inconsistent Results After ONNX Runtime Optimization

  #23199 opened Dec 26, 2024

• PyExc_Exception while import onnxruntime

  #23196 opened Dec 26, 2024

• [Mobile] How to use GPU acceleration on Android

  #23194 opened Dec 25, 2024

• [Build] TypeInferenceError when quantize an onnx model with custom operator

  #23191 opened Dec 25, 2024

• [Feature Request] Shape inference for GroupQueryAttention Op

  #23189 opened Dec 24, 2024

• Error merging shape info for output

  #23187 opened Dec 24, 2024

• [Web] Upgrading from 1.20.1 to 1.21.* breaks Segment Anything models on WebGPU

  #23183 opened Dec 23, 2024

• [Build] Fails on arm64: error: no member named 'linux_id' in 'cpuinfo_processor'

  #23181 opened Dec 23, 2024

• [Build] error: array index 7 is past the end of the array (that has type '__m256[4]')

  #23180 opened Dec 23, 2024

• Cmake Build Failed due to undefined symbol: MlasGemmX8S8DispatchNeon

  #23177 opened Dec 22, 2024

28 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• [Fix] in Xnnpack EP, the conversion for fused activation param isn't correct

  #23115 commented on Dec 26, 2024 • 8 new comments

• [QNN EP] [DRAFT] Make QNN EP a shared library

  #23120 commented on Dec 23, 2024 • 4 new comments

• [webgpu] support Pad operator

  #23141 commented on Dec 27, 2024 • 2 new comments

• [CoreML] support coreml model cache

  #23065 commented on Dec 24, 2024 • 1 new comment

• Adding optional CUDA DLLs when installing onnxruntime_gpu

  #22506 commented on Dec 22, 2024 • 1 new comment

• [js/node] allow arenaExtendStrategy and gpuMemLimit option for CUDA EP

  #23176 commented on Dec 23, 2024 • 0 new comments

• Update onnxruntime_c_api.h to work with MinGW

  #23169 commented on Dec 23, 2024 • 0 new comments

• [WIP] Stable Diffusion 3.x and Flux Optimization

  #22986 commented on Dec 27, 2024 • 0 new comments

• Bump onnx from 1.16.1 to 1.17.0 in /onnxruntime/python/tools/transformers/models/phi2

  #22928 commented on Dec 24, 2024 • 0 new comments

• [ROCm] add -Wno-pass-failed for CK build

  #20859 commented on Dec 26, 2024 • 0 new comments

• [Build]

  #13606 commented on Dec 28, 2024 • 0 new comments

• [Build] Float16_t and BFloat16_t compile error

  #20564 commented on Dec 27, 2024 • 0 new comments

• [Web] BiRefNet_T not working on webgpu

  #21968 commented on Dec 27, 2024 • 0 new comments

• [Build] Build Error

  #22882 commented on Dec 26, 2024 • 0 new comments

• [Performance] how to set the threads when using TRT EP

  #22913 commented on Dec 26, 2024 • 0 new comments

• [Performance] Observing higher memory spikes in C++ when running multiple Inference `Run()` executions on CPU

  #22920 commented on Dec 26, 2024 • 0 new comments

• [DML EP] ORT would crash after deleting one of the models and then doing an inference

  #22948 commented on Dec 26, 2024 • 0 new comments

• [WebGPU] `Error: [WebGPU] Kernel "[Mul] /head/istft/Mul_1" failed. Error: Failed to generate kernel's output[0] with dims [1,3520,3520]. If you are running with pre-allocated output, please make sure the output type/dims are correct. Error: 81415528.`

  #22994 commented on Dec 26, 2024 • 0 new comments

• [Build] passing --arm64 to ci_build/build.py has error in arm64 host

  #20814 commented on Dec 25, 2024 • 0 new comments

• [Build] Unable to cross-compile ONNX Runtime 1.17.1 for ARM Cortex A53

  #23152 commented on Dec 24, 2024 • 0 new comments

• [Performance] Inference failed or unsupported using quantize_dynamic

  #20060 commented on Dec 24, 2024 • 0 new comments

• Discrepancies in ONNX Runtime Inference Results on RISC-V

  #22530 commented on Dec 24, 2024 • 0 new comments

• About NVIDIA Jetson TX1/TX2/Nano/Xavier/Orin Builds

  #23113 commented on Dec 24, 2024 • 0 new comments

• [Build] Building onnxruntime with version v1.18.2 is successful but gives an error with v1.20.1, cannot understand what I am doing wrong

  #23150 commented on Dec 23, 2024 • 0 new comments

• [Feature Request] Support pointer-generator networks on T5 BeamSearch

  #23122 commented on Dec 23, 2024 • 0 new comments

• Very close on this. I believe the majority of this code is setup correctly. cudaErrorInvalidValue:invalid argument

  #23149 commented on Dec 23, 2024 • 0 new comments

• [Build] build error for windows

  #23166 commented on Dec 23, 2024 • 0 new comments

• [Build] onnxruntime_gpu PiPy on a slow host

  #23079 commented on Dec 22, 2024 • 0 new comments