# Keras

Deep Learning for humans

Repository: https://github.com/keras-team/keras
Canonical: https://ross.abutalabs.com/products/keras
Homepage: http://keras.io/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deep-learning, tensorflow, neural-networks, machine-learning, data-science, python, jax, pytorch
Last push: 2026-08-26T22:12:59+00:00

## Health v2 (maintenance only)
Score: 98/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 95, longevity 100
- inputs: {"age_days": 4177, "days_push": 7, "days_rel": 35, "gap_med": 30.0, "n_releases_24m": 23}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 64253, forks 19764 (observed 2026-08-28T04:12:19.205514+00:00)

## What it is
Keras 3 is a multi-backend deep learning framework that runs on JAX, TensorFlow, PyTorch, and OpenVINO, providing a high-level API for building and training neural networks. It supports workloads from laptops to large GPU/TPU clusters across computer vision, NLP, audio, and timeseries tasks.

## Use cases
- build and train neural networks in python
- train deep learning models on gpu or tpu clusters
- switch between tensorflow pytorch and jax backends
- build computer vision or nlp models with a high-level api
- forecast timeseries with deep learning
- run model inference with openvino

## When to choose
- you want a simple, high-level API for deep learning in Python
- you need to train models across multiple hardware backends
- you want to scale from a laptop to datacenter GPU/TPU clusters
- you prefer easy-to-debug eager execution with PyTorch or JAX

## When to avoid
- you need low-level control over training loops and graph internals
- you work primarily on Windows without WSL2
- you need inference-only deployment without a backend framework

## Facets
- artifact type: library
- maturity: stable
- function: deep-learning, machine-learning, llm-training, data-science
- domain: deep-learning, machine-learning, data-science, computer-vision
- platform: python
- tags: neural-networks, multi-backend, tensorflow, pytorch, jax, openvino, model-training, high-level-api, natural-language-processing, linux, macos, gpu

## Member repositories
- keras-team/keras (main) score 98
- keras-team/keras-core (mirror) score 10
- keras-team/keras-preprocessing (mirror) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:19.205514+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 2026-08-29T16:17:25.681009+00:00, confidence not recorded.
  - readme: https://github.com/keras-team/keras (fetched 2026-08-28T04:12:19.205514+00:00, sha db4fa3c4a3e9)
- Data as of 2026-08-30T08:39:29.467469+00:00.