5 Pull requests merged by 4 people

• Enable float8e8m0 for Q/DQ, and other ops

  #7120 merged Jul 12, 2025

• Test ArgMax with multiple maximal values and select_last_index = 0

  #7104 merged Jul 10, 2025

• Use specific version for mac os runner

  #7115 merged Jul 10, 2025

• Add FLOAT8E8M0 data type

  #7030 merged Jul 10, 2025

• Fix cast test cases

  #7102 merged Jul 8, 2025

4 Pull requests opened by 4 people

• Add TensorScatter op for in-place kv cache udpate

  #7114 opened Jul 9, 2025

• Add more test cases with attention

  #7117 opened Jul 10, 2025

• Update CMakeLists.txt to prevent ICE protobuf failure

  #7121 opened Jul 12, 2025

• Add context to ONNX checker ValidationError messages for better debugging

  #7122 opened Jul 12, 2025

5 Issues closed by 1 person

• Print model initializers with `onnx.printer`

  #5104 closed Jul 12, 2025

• macos-runner? 14 or 15?

  #7113 closed Jul 10, 2025

• onnx.shape_inference.infer_shapes creates a blank graph when passed a >2GB model

  #6150 closed Jul 8, 2025

• Error While Installing ONNX

  #6140 closed Jul 7, 2025

• A conflict doc abount compatibility between Onnx Version & ML Opset Version

  #6134 closed Jul 7, 2025

2 Issues opened by 2 people

• [ONNX] Elu operator should support input tensor with any shape rather than only 1D

  #7119 opened Jul 11, 2025

• [ONNX][Attention]the reference missed Transpose for 3D attention

  #7118 opened Jul 11, 2025

18 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.

• Arguments misaligned with nodes

  #7101 commented on Jul 7, 2025 • 0 new comments

• Export onnx model from pytorch with dynamic input size, output tensor doesn't match.

  #6141 commented on Jul 8, 2025 • 0 new comments

• Discussion on the Correct Approach for Converting ONNX Models to Other Frameworks

  #6142 commented on Jul 8, 2025 • 0 new comments

• OperatorSetProto.onnx: the instance of immutable operator set domain ai.onnx and ai.onnx.ml

  #6163 commented on Jul 8, 2025 • 0 new comments

• Expected output of QLinearMatMul: using astype or np.clip?

  #6169 commented on Jul 9, 2025 • 0 new comments

• The model is converted to onnx format using dynamic batch precision collapse

  #6192 commented on Jul 9, 2025 • 0 new comments

• [Feature request] ONNX need support opset MatMulInteger input 4bit

  #6547 commented on Jul 9, 2025 • 0 new comments

• Tensor and Integer Comparison Problem in ONNX Export

  #6212 commented on Jul 10, 2025 • 0 new comments

• Is TfIdfVectorizer safe to use?

  #6078 commented on Jul 11, 2025 • 0 new comments

• Segfault in Einsum shape inference

  #6981 commented on Jul 11, 2025 • 0 new comments

• Create error codes and reference pages for checker errors

  #6082 commented on Jul 13, 2025 • 0 new comments

• ONNX checker missing context

  #6085 commented on Jul 13, 2025 • 0 new comments

• DequantizeLinear spec clarification: What happens if the subtraction overflows/underflows?

  #6132 commented on Jul 14, 2025 • 0 new comments

• [Feature request] Better covered and scalable backend tests

  #5056 commented on Jul 14, 2025 • 0 new comments

• Fix Einsum shape inference segfault for scalar inputs

  #7055 commented on Jul 7, 2025 • 0 new comments

• Additional test cases for 'MatMul' operator

  #7103 commented on Jul 10, 2025 • 0 new comments

• Use main protobuf to temporarily fix Windows builds

  #7110 commented on Jul 7, 2025 • 0 new comments

• Add support for constructing functions with graph attributes

  #7112 commented on Jul 7, 2025 • 0 new comments