AI Backends

AI Backends is a self-hosted, open-source API server that exposes a consistent REST interface to multiple large language model providers and local models. It provides ready-made endpoints for common AI tasks (summarization, translation, sentiment, image analysis, Q&A, meeting notes, project planning and more) so developers can integrate AI features without changing client code when switching providers. The project is distributed as source code that you clone and run locally or in your own infrastructure, and it includes configuration to route requests to OpenAI, Ollama, Anthropic, OpenRouter, LMStudio, Vercel AI Gateway and other providers.

• Unified API endpoints — Use built-in REST endpoints like /api/summarize, /api/translate, /api/sentiment, /api/askText, /api/describeImage and more to add common AI capabilities quickly.
• Multiple provider support — Configure providers (OpenAI, Ollama, Anthropic, OpenRouter, LMStudio, Vercel AI Gateway) in a single configuration and switch providers without code changes.
• Self-hosted & open-source — Clone the repository and run the server locally or in your environment to retain full control over data and deployment.
• Extensible configuration — Provide provider-specific model, function-calling and request configuration in requests to tailor behavior per provider.
• Developer friendly quick start — Clone the repository, install dependencies, and run the server to get a local API at http://localhost:3000 for development and testing.

To get started, clone the official repository, follow the included quick-start commands, and update the provider configuration to point to your chosen model providers or local inference runtime.