Hugging Face integration

Run Machine Learning models from Hugging Face in three clicks

Use Livebook's Neural Network Smart cell to download and run Machine Learning models from Hugging Face without a single line of code. Customize the generated code to your needs.

Install Livebook Locally or on the cloud

How to Build an ML app

Text to image

Generate images using the Stable Diffusion model.

Image classification

Classify images using models like ResNet, ConvNeXT, ViT, and DeiT.

Text classification

Classify text using models like RoBERTa, DistilRoBERTa, and FinBERT.

Token classification

Understand the structure of a text via named-entity recognition.

Zero-shot text classification

Classify text using your own labels with models like BART and XLM-RoBERTa multilingual.

Fill-mask

Predict which words should replaces masks inside text.

Text generation

Generate text using models like GPT2, DistilGPT2, and BART.

Speech-to-text

Generate text from audio using the Whisper model.

Use the generated code inside your app

You can see the code generated by the Smart cell to understand how a model works and embed it directly into your Elixir app.

Run Livebook inside Hugging Face Spaces

You can run Livebook in Hugging Face Spaces and leverage their GPU offerings to train and run Machine Learning models faster and more efficiently.