# transcranial/keras-js

Run Keras models in the browser, with GPU support using WebGL

Repository: https://github.com/transcranial/keras-js
Canonical: https://ross.abutalabs.com/products/keras-js
Homepage: https://transcranial.github.io/keras-js
Language: JavaScript
License: MIT
License Family: permissive
Topics: deep-learning, machine-learning, webgl, tensorflow, neural-networks, keras, javascript
Last push: 2022-06-15T12:11:24+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3624, "days_push": 1540, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4963, forks 491 (observed 2026-08-28T04:09:03.939593+00:00)

## What it is
A JavaScript library for running trained Keras deep learning models directly in the browser, with GPU acceleration via WebGL 2. It also supports CPU-only execution in Node.js and works with models trained on any Keras backend such as TensorFlow or CNTK.

## Use cases
- run keras models in the browser
- client-side neural network inference with webgl
- deploy mnist cnn classifier in a web page
- run resnet50 or inception v3 image classification in javascript
- browser-based lstm sentiment classification
- execute deep learning models without a server

## When to choose
- you have legacy Keras 2.1.2 models and need browser inference without migrating
- you want to study how Keras models are serialized and executed in JavaScript

## When to avoid
- starting any new project - the project is officially inactive and recommends TensorFlow.js
- you need support for recent Keras versions or modern model architectures
- you need ongoing maintenance, bug fixes, or community support

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, deep-learning, llm-inference
- domain: deep-learning, machine-learning, web-development, browser-extensions
- platform: browser
- tags: keras, webgl, gpu, inference, neural-networks, tensorflow, deprecated, nodejs, javascript

## Member repositories
- transcranial/keras-js (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:03.939593+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:17:47.278990+00:00, confidence not recorded.
  - readme: https://github.com/transcranial/keras-js (fetched 2026-08-28T04:09:03.939593+00:00, sha b5b19310f97c)
  - homepage: https://transcranial.github.io/keras-js (fetched 2026-08-29T08:59:27.423852+00:00, sha ec583e57d003)
  - registry_npm: https://registry.npmjs.org/keras-js (fetched 2026-08-29T08:59:27.432769+00:00, sha 7508c9af8da6)
- Data as of 2026-08-30T08:39:29.467469+00:00.