Lemonade AI

Lemonade provides a local AI runtime for text, images, and speech that can run on user devices. It aims to be free to use, private by design, and adaptable by a community of builders. The project emphasizes openness and a local-first approach, where data stays on the user’s device and no telemetry is reported. It supports chat, vision, image, speech, transcription, and embeddings within apps or agents.

The platform offers installation options that bring an AI runtime, a graphical user interface, a command-line interface, and API endpoints to the host system. It highlights cross-platform availability and the ability to run on Windows, macOS, and various Linux distributions, with packaging tools like Docker and Snap mentioned.

Lemonade is designed to work with many apps and is compatible with the OpenAI API standard, enabling integration with hundreds of apps. The project releases updates weekly and maintains a public roadmap and a marketplace for integrations. The community around Lemonade includes industry veterans and developers who contribute to its open-source engines and ecosystem.

The Embeddable Lemonade guide describes a portable build that can be bundled into an app, detailing file layouts, customization options, and deployment steps. It explains how to configure settings, pre-download backends, and place the embeddable folder into an application's installer, with guidance on preserving a private Lemonade instance and per-model customization.

Use cases

• Embeddable Lemonade is a portable build of the lemond service that you can bundle into your app.