OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Developer Tools & SDKs

Keras

Keras 3 is an open-source Python deep-learning framework that lets developers build and train neural-network models through a high-level interface over JAX, TensorFlow and PyTorch, with OpenVINO available for inference-only workflows.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Keras — OpenViST project profile in AI Developer Tools & SDKs.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Keras when a project needs a high-level modeling interface with a choice of JAX, TensorFlow or PyTorch backends, or when OpenVINO inference is required.

Who it is for

  • Python developers building deep-learning models
  • Machine-learning teams working with multiple supported backends
  • Developers creating computer-vision, natural-language, audio, forecasting or recommendation models

Documented uses

  • Build and train neural-network models
  • Develop models across JAX, TensorFlow and PyTorch
  • Create inference applications with OpenVINO
  • Reuse Keras models and components in framework-specific workflows
  • Adapt existing tf.keras projects to Keras 3
Before you choose

Requirements & limits

The public repository identifies Keras as open source under the Apache-2.0 license. The documented installation uses Python package management and requires a separately installed backend.

  • Choose JAX, TensorFlow or PyTorch for model development and training; OpenVINO is inference-only.
  • Configure the backend before importing Keras because it cannot be changed afterward.
  • GPU workflows require a pre-installed NVIDIA driver and backend-specific CUDA dependencies.
  • The documented local installation is compatible with Linux and macOS; Windows users are directed toward WSL2.
  • Existing tf.keras projects may require updates to save models in .keras format.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Keras and configure a backend

Set up a local Keras installation with a selected supported backend.

Before you start

  • Python environment
  • Keras 3 package
  • One backend package: TensorFlow, JAX or PyTorch
  • Linux or macOS; Windows development is recommended through WSL2

Steps

  1. Install the keras package with pip.

  2. Install the selected backend package.

  3. Set KERAS_BACKEND or edit the local Keras configuration file.

  4. Configure the backend before importing keras.

Expected outcome

A local Keras installation configured to use the selected backend.

Guide limitations

  • OpenVINO is available for inference-only workflows.
  • The backend cannot be changed after Keras has been imported.
  • The documented minimum backend versions vary by latest stable Keras 3 release.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
1 October 2026
Original website title
keras-team/keras · Official homepage title · retained in its source language
How we review sourcesSuggest a correction