Keras 3: Deep Learning for Humans by keras-team
Deep Learning for humans
About Keras 3: Deep Learning for Humans
From the project's README at github.com/keras-team/keras. Lightly cleaned for readability; for the full source see the upstream repo.
Keras 3 is a multi-backend deep learning framework, with support for JAX, TensorFlow, PyTorch, and OpenVINO (for inference-only). Effortlessly build and train models for computer vision, natural language processing, audio processing, timeseries forecasting, recommender systems, etc. Accelerated model development: Ship deep learning solutions faster thanks to the high-level UX of Keras and the availability of easy-to-debug runtimes like PyTorch or JAX eager execution. State-of-the-art performance: By picking the backend that is the fastest for your model architecture (often JAX!), leverage speedups ranging from 20% to 350% compared to other frameworks. Benchmark here. Datacenter-scale training: Scale confidently from your laptop to large clusters of GPUs or TPUs.
Join nearly three million developers, from burgeoning startups to global enterprises, in harnessing the power of Keras 3. Installation Install with pip
Keras 3 is available on PyPI as . Note that Keras 2 remains available as the package.
To use , you should also install the backend of choice: , , or . Additionally, The backend is available with support for model inference only. Local installation Minimal installation
Health score breakdown
6-dimension composite. See methodology for formula and weights.
Adoption signals
Real-world usage data, pulled from each registry. The bigger the numbers, the more battle-tested the project.
| Signal | Value | Source |
|---|---|---|
| GitHub stars | 64k | github.com/keras-team/keras |
| GitHub forks | 20k | github.com/keras-team/keras |
Release & maintenance
Is this project actively maintained, or about to die? Check the recency of last commit and last release.
| Project age | 11.1 years | since Mar 2015 |
| Last commit | 2 days ago | May 5, 2026 |
| Releases shipped | 87 | last: 1 months ago |
| Security policy | SECURITY.md | declared by maintainers |
Self-hosting cost across providers
Detected requirements: 4GB RAM, 40GB disk minimum. Cheapest plan per provider that meets the requirement.
| Provider | Plan | Specs | Monthly | |
|---|---|---|---|---|
| hetzner | CAX11 | 2c · 4GB · 40GB | $4.13 USD | Deploy → |
| vultr | VC2 | 1c · 1GB · 25GB | $5 USD | Deploy → |
| linode | Nanode 1GB | 1c · 1GB · 25GB | $5.12 USD | Deploy → |
| digitalocean | Basic Regular 1GB | 1c · 1GB · 25GB | $6 USD | Deploy → |
Security advisories
GHSA-mgx6-5cf9-rr43. What people say on Hacker News
Ready to self-host Keras 3: Deep Learning for Humans?
Spin up a hetzner CAX11 (4GB RAM, 40GB disk) for $4.13/mo and follow the project's official install docs.
Data last refreshed May 7, 2026.
Similar open-source projects
Projects in our directory that replace the same SaaS or share topics with Keras 3: Deep Learning for Humans.