Show HN: Tynx - train ONNX models with a PyTorch-shaped API (<20 MB whl)

Tynx, a new ONNX runtime with a PyTorch-shaped API for inference and training, has been released as a Python package under 20 MB. Written in Rust, it uses Burn/CubeCL and wgpu for GPU execution across operating systems and GPUs without extra libraries, and can compile to the browser. The project is early-stage, seeking feedback and use cases.

Tynx is a small, self-contained ONNX runtime with a PyTorch-shaped API for inference and training. The .whl is less than 20MB. GPU execution uses Burn/CubeCL + wgpu enabling it to run across OSes and GPUs, without requiring any extra libraries. bash $ pip install tynx And API python import tynx as tx model = tx.nn.Sequential tx.nn.Linear 8, 16 , tx.nn.ReLU , tx.nn.Linear 16, 2 optimizer = tx.optim.Adam model.parameters , lr=1e-3 loss = tx.nn.functional.cross entropy model x , target loss.backward optimizer.step The runtime is written in Rust, and also can compile to the browser PoC done .It’s early, I'd love for feedback and possible use cases & API expansion where this could be beneficial. Comments URL: https://news.ycombinator.com/item?id=49268144 https://news.ycombinator.com/item?id=49268144 Points: 1 Comments: 0