sonos/tract
Tiny, no-nonsense, self-contained, Tensorflow and ONNX inference
This is a neural network inference toolkit that helps you run trained AI models in production environments. It takes models saved in formats like ONNX or NNEF and efficiently executes them, providing predictions or classifications. It's designed for engineers deploying AI models into applications, especially on resource-constrained devices, or those looking to optimize model performance.
2,818 stars. Actively maintained with 237 commits in the last 30 days.
Use this if you need to run pre-trained machine learning models quickly and efficiently in an application, converting models from training frameworks like TensorFlow or PyTorch into a deployable format.
Not ideal if you are looking for a tool to train neural networks or if your models heavily rely on advanced ONNX features like Tensor Sequences and Optional Tensors.
Stars
2,818
Forks
250
Language
Rust
License
—
Category
Last pushed
Mar 13, 2026
Commits (30d)
237
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/sonos/tract"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
tracel-ai/burn
Burn is a next generation tensor library and Deep Learning Framework that doesn't compromise on...
pykeio/ort
Fast ML inference & training for ONNX models in Rust
elixir-nx/ortex
ONNX Runtime bindings for Elixir
robertknight/rten
ONNX neural network inference engine
mosure/burn_dragon
burn inference and training of baby dragon hatchling 🔥🐉