{"slug": "open-sourcing-the-atuin-ai-server", "title": "Open Sourcing the Atuin AI Server", "summary": "Atuin has open-sourced the server for its terminal-based AI agent, allowing users to self-host the service for greater data privacy. The Atuin AI server supports any OpenAI-compatible endpoint, including local models like Ollama and vLLM, and can be run from source or via Docker. This move addresses user concerns about terminal data privacy while maintaining the agent's functionality.", "body_md": "# Open Sourcing the Atuin AI Server\n\nSelf host your own server for Atuin AI - the agent that lives in your terminal, and knows how you work\n\nAtuin AI is a fast terminal-focused agent right in your shell. It starts instantly and comes with all the agentic tools needed to help with work in your terminal.\n\nAtuin offers a generous usage budget for free for all users, and we don't store your AI conversations, but we recognize some people would prefer to keep tighter tabs on their data, especially data that originates from terminal usage.\n\nTo that end, we're happy to be [open sourcing the Atuin AI server](https://github.com/atuinsh/atuin-ai-server?ref=blog.atuin.sh).\n\n## The Atuin AI Server\n\nThe Atuin AI server can be found on GitHub at [atuinsh/atuin-ai-server](https://github.com/atuinsh/atuin-ai-server?ref=blog.atuin.sh); it's based on [atuinsh/atuin-ai-core](https://github.com/atuinsh/atuin-ai-core?ref=blog.atuin.sh), the same library we use for the production Atuin AI server.\n\nThe Atuin AI server currently supports any OpenAI-compatible, chat completions-style endpoint. For local models, this includes Ollama, vLLM, LM Studio, llama.cpp, and LiteLLM, among others. You can also use OpenAI-compatible web services, like OpenRouter.\n\n## Getting Started\n\nAfter cloning the repository, copy the example config file, `config.example.toml`\n\n, to `config.toml`\n\n. Follow the configuration section of the readme to set up your instance.\n\nHere's a very basic example of an Ollama-based setup:\n\n```\nport = 8080\nendpoint = \"http://localhost:11434/v1\" # or host.docker.internal\napi_key = \"ollama\"\n\ndefault_model = \"llama31\"\n\n[request.body]\nstream_options = { include_usage = true }\n\n[[models]]\nalias = \"llama31\"\nname = \"Llama 3.1 70b\"\ndescription = \"Ollama Llama 3.1 70b\"\nmodel = \"llama3.1:70b\"\n\n[[models]]\nalias = \"gemma4\"\nname = \"Gemma 4 r4b\"\ndescription = \"Ollama Gemma 4 - Effective 4b\"\nmodel = \"gemma4:e4b\"\n```\n\nSee the [repository readme](https://github.com/atuinsh/atuin-ai-server?ref=blog.atuin.sh#readme) for more setup details, including configuring server-side tools, like web search and web content scraping.\n\nOnce done, you can start the server one of two ways:\n\n### Running from Source\n\nIf you have Erlang, Elixir, and Gleam installed, you can run the server natively:\n\n```\nmix deps.get\nmix run --no-halt\n```\n\nIf your `config.toml`\n\nspecifies API keys via environment variables, remember to set them when you start the server.\n\n### Running with Docker\n\nTo run the server with docker, run the following:\n\n```\ndocker run \\\n  -v ./config.toml:/etc/atuin-ai/config.toml \\\n  -p 8080:8080 \\\n  ghcr.io/atuinsh/atuin-ai-server:latest\n```\n\nIf you're running via Docker and want the Atuin AI server to connect with a local LLM service running on the host, like Ollama, use `host.docker.internal`\n\nas the endpoint instead of `localhost`\n\n(which would resolve to the container's own loopback interface).\n\n## Configuring Atuin AI\n\nOnce your server is running, you can configure Atuin AI to connect to it by setting [the endpoint config](https://docs.atuin.sh/cli/ai/settings/?ref=blog.atuin.sh#endpoint):\n\n```\n[ai]\nendpoint = \"http://localhost:8080\"\n```\n\n", "url": "https://wpnews.pro/news/open-sourcing-the-atuin-ai-server", "canonical_source": "https://blog.atuin.sh/atuin-ai-oss/", "published_at": "2026-07-14 18:21:23+00:00", "updated_at": "2026-07-14 18:48:27.881520+00:00", "lang": "en", "topics": ["ai-tools", "ai-infrastructure", "developer-tools"], "entities": ["Atuin", "Ollama", "vLLM", "LM Studio", "LiteLLM", "OpenRouter", "GitHub"], "alternates": {"html": "https://wpnews.pro/news/open-sourcing-the-atuin-ai-server", "markdown": "https://wpnews.pro/news/open-sourcing-the-atuin-ai-server.md", "text": "https://wpnews.pro/news/open-sourcing-the-atuin-ai-server.txt", "jsonld": "https://wpnews.pro/news/open-sourcing-the-atuin-ai-server.jsonld"}}