google-deepmind/sonnet

TensorFlow-based neural network library

58
/ 100
Established

Sonnet is for machine learning researchers and practitioners who want to build custom neural networks using TensorFlow. It provides a flexible way to define neural network components (modules) and combine them, taking raw data (like images or text) as input and producing model predictions or learned representations as output. This allows for deep customization of neural network architectures.

9,907 stars.

Use this if you are a machine learning researcher or engineer who needs to design and implement novel neural network architectures with precise control over their components and behavior within the TensorFlow 2 ecosystem.

Not ideal if you are looking for an all-in-one machine learning framework with built-in training loops, data pipelines, and high-level APIs for common model types, as Sonnet focuses specifically on neural network construction.

neural-network-design deep-learning-research custom-model-building artificial-intelligence machine-learning-engineering
No Package No Dependents
Maintenance 10 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

9,907

Forks

1,304

Language

Python

License

Apache-2.0

Last pushed

Feb 10, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/google-deepmind/sonnet"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.