Welcome
Run, fine-tune and deploy open-source AI models with the Together AI REST API.
The Together AI API lets you run inference on hundreds of open-source models (chat, completions, embeddings, images, audio, rerank), fine-tune them on your own data, and deploy them on dedicated endpoints. It is OpenAI-compatible, so existing clients work by changing the base URL.
What you'll find here
- API Reference: all 201 operations grouped by capability (Chat, Completions, Embeddings, Images, Audio, Fine-tuning, Files, Batches, Endpoints, Deployments and more), with request and response examples and a live API playground.
- SDKs: client libraries for C#, Python and TypeScript, generated by APIMatic, with code samples on every endpoint.
- Context Plugin: brings the API and SDKs into AI coding assistants such as Claude Code, Cursor and Codex.
Base URL
https://api.together.ai/v1Authentication
Send your API key as a bearer token:
Authorization: Bearer $TOGETHER_API_KEYCreate a key in your Together AI settings.
Common tasks
- Chat with a model:
POST /chat/completions - Create embeddings:
POST /embeddings - Generate an image:
POST /images/generations - List available models:
GET /models - Start a fine-tuning job:
POST /fine-tunes