tensorflow/tfjs

Health v2 · maintenance only

Adoption not part of the score

19134 stars · 2030 forks observed · 2026-08-28

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

TensorFlow.js is a hardware-accelerated JavaScript library for training and deploying machine learning models in the browser and Node.js. It provides low-level and Keras-like high-level APIs, multiple backends (CPU, WebGL, WASM, WebGPU, native Node), and converters for running pre-trained TensorFlow models in JavaScript.

Use cases

• run machine learning models in the browser
• train neural networks in javascript
• convert a python tensorflow model to run on the web
• do transfer learning with webcam data in the browser
• run inference client-side without a server
• build an ml-powered web app with webgl acceleration
• use pretrained models like object detection in javascript

When to choose

• you need ML inference or training directly in the browser with GPU acceleration
• you want to run or fine-tune TensorFlow models from JavaScript or Node.js
• you need client-side privacy by keeping data and models on the device
• you want a Keras-like layers API in TypeScript/JavaScript

When to avoid

• you are training large-scale models where Python TensorFlow or PyTorch is more appropriate
• you need maximum native GPU performance for heavy training workloads
• your project has no JavaScript runtime and is purely server-side Python

Facets

library · maturity stable

machine-learning deep-learning gpu-computing data-visualization machine-learning deep-learning web-development frontend browser cross-platform wasm tensorflow webgl webgpu neural-networks model-conversion transfer-learning pretrained-models typescript nodejs web gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="tensorflow/tfjs")