BrainJS/brain.js

Health v2 · maintenance only

Adoption not part of the score

14861 stars · 1077 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

brain.js is a GPU-accelerated neural network library written in JavaScript/TypeScript that runs in both browsers and Node.js. It supports feed-forward, recurrent (RNN, LSTM, GRU), and convolutional networks with a simple, easy-to-use API.

Use cases

• train a neural network in the browser
• run machine learning in node.js without python
• predict time series values with LSTM in javascript
• build a simple classifier with a feed-forward network
• gpu accelerated neural network training in javascript
• forecast future values from historical data in the browser

When to choose

• you want neural networks in JavaScript for browser or Node.js without a Python toolchain
• you need a simple, beginner-friendly API for feed-forward, RNN, LSTM, or GRU networks
• you want GPU acceleration for training in the browser via WebGL

When to avoid

• you need large-scale deep learning with modern architectures like transformers
• you require an extensive ecosystem of ML tooling, model zoos, and research implementations
• you need production-grade training performance comparable to PyTorch or TensorFlow

Facets

library · maturity active

machine-learning deep-learning machine-learning deep-learning artificial-intelligence browser cross-platform neural-networks gpu-accelerated lstm rnn time-series javascript typescript nodejs

Member repositories

For agents

markdown · JSON · MCP: product_card(name="BrainJS/brain.js")