jax2onnx-models

🧪 This repository hosts ONNX models generated by the jax2onnx converter as part of its automated pytest suite.

These models are used to:

• Validate correctness of the ONNX export pipeline
• Provide public artifacts for visualization and inspection (e.g., with Netron)
• Enable CI verification of shape/type fidelity, runtime compatibility, and graph structure

---

📦 Format

All models are stored in the ONNX (Open Neural Network Exchange) format.
They are compatible with:

• ONNX runtimes (ONNX Runtime, TensorRT, etc.)
• Netron for visual graph inspection
• External ONNX validators

Models may include both small unit test graphs and larger architectures like GPT LLMs or Vision Transformers.

---

🔁 Update Policy

• This repository is updated regularly with models generated during testing.
• Only the latest model state is kept — older versions are overwritten as needed.
• We do not maintain file history to keep the repository lightweight.

---

📄 License

This repository and all included model files are licensed under the Apache License 2.0.

See LICENSE for details.