Self-host Big-AGI
v2.1.3Multi-model AI workspace with parallel Beam runs
docker run -d -p 3000:3000 ghcr.io/enricoros/big-agi:latest- Stars
- 7 166
- Port
- 3000 (HTTP web app (Next.js server); run/compose map it as 3000:3000)
ghcr.io/enricoros/big-agiWhat is Big-AGI?
Big-AGI Open is the MIT-licensed codebase behind Big-AGI, a browser-based AI workspace aimed at engineers, researchers and IT teams who work with many frontier models at once. Its signature capability, Beam, fires a prompt at between two and twenty-four models in parallel and then fuses the replies into one answer, so agreement and disagreement both become visible. Personas bundle instructions, documents, images, memory, a chosen model and voice settings into reusable profiles, while the AI Inspector exposes each outgoing request with its parameters, tokens and cost. Model access is bring-your-own-key across twenty-plus cloud vendors plus local servers such as Ollama, LocalAI and LM Studio, alongside image generation, speech synthesis and voice calls, web search with citations, web-page download, file and PDF ingestion, split-pane multi-chat, chat export and cost estimates. Conversations stay on the device by default; a database (Postgres or MongoDB) is only needed for shared chat links, and the vendor also sells a $10.99/month hosted Pro plan that adds cross-device sync and storage. Self-hosting is a single Docker container listening on port 3000, with Docker Compose, Kubernetes manifests or a Vercel fork as alternatives; user authentication is not built in, so protected deployments rely on optional HTTP Basic (requires a manual build) or a cloud platform / reverse proxy in front.
Should you self-host Big-AGI?
Big-AGI scores 67/100 on selfhost.directory (how scoring works) and ranks #10 of 68 in AI Chat. It has 7 166 GitHub stars, +29 in the last 30 days.
It shipped 9 stable releases in the past year. The latest, v2.1.3, came out on 27 October 2026. New versions land roughly every 2 months.
Setup is easy: most people get it running in one sitting. An official Docker image is available. It is MIT-licensed and written mostly in Typescript.
Its closest alternative on selfhost.directory is LibreChat (85/100, 46 844 stars). It scores higher overall, so compare the two before you pick. See Big-AGI vs LibreChat side by side.
Features
- AI answer merging
- AI personas with attached context
- AI request inspector
- Bring your own API keys
- Chat cost estimation
- Chat export
- Cross-device cloud syncHosted big-agi.com Pro tier, not part of the self-hosted open build.
- File attachments and PDF extraction
- Image generation and editing
- Local model server support
- Multi-model parallel chat (Beam)
- Split-pane multi-chat
- Voice calls and speech synthesis
- Web page download
- Web search with citations
Big-AGI README
This is the open-source foundation of Big-AGI, the multi-model AI workspace for experts.
Big-AGI is the multi-model AI workspace for experts: Engineers architecting systems. Founders making decisions. Researchers validating hypotheses. You need to think broader, decide faster, and build with confidence, then you need Big-AGI.
It comes packed with world-class features like Beam, and is praised for its best-in-class AI chat UX. As an independent, non-VC-funded project, Pro subscriptions at $10.99/mo fund development for everyone, including the free and open-source tiers.
FAQ
How do I verify my Vercel deployment version?
Open the deployments section of your Vercel project to see the latest status and time; each deployment links to the source commit it was built from.
How do I check my Big-AGI version?
The running version is shown in the News section inside the app.
What is "Direct Connection" and should I enable it?
It lets the browser call the AI provider directly instead of going through the app's edge server, removing the upload-size and timeout caps and improving privacy. It is unavailable when keys are stored server-side or when the provider blocks cross-origin requests.
More in AI Chat
bionic-gptBionic is an open-source, self-hosted agentic AI platform for internal teams to build sovereign workflows with own models and data — deployable on-premise.
chatbot-ollamaSelf-hosted, open-source chat web UI for Ollama: stream local model replies, upload documents and images, and switch between dark and light themes.
chatbot-uiOpen-source AI chat app you self-host with a Supabase backend, supporting OpenAI, Anthropic, Gemini, Mistral and Azure keys plus local Ollama models.
ChatbotXSelf-hosted, open-source ManyChat alternative for WhatsApp, Messenger, Instagram, TikTok and webchat: visual flow builder, AI agents, live inbox, CRM and broadcasts.
ChatChatSelf-hosted, open-source unified AI chat and search platform with one interface over OpenAI, Anthropic, Cohere, Google and more, deployed via Docker, Vercel or Railway.
ChatGPT Next WebSelf-hosted, open-source AI chat UI for OpenAI, Claude, Gemini, DeepSeek and many other providers, with Docker, Vercel and native desktop/mobile clients.
chatgpt-uiSelf-hosted ChatGPT web client with multi-user support, multiple languages, and persistent storage (MySQL/PostgreSQL/SQLite) via Nuxt frontend and Django backend.
chatGPT-webOpen-source, self-hosted single-page chat front end for OpenAI-compatible APIs: pick a provider, paste your key, and chats stay in browser storage.
ChatGPT Web Midjourney ProxySelf-hosted open-source web UI that combines ChatGPT, Midjourney, Suno, Luma, Runway, Pika, and other AI models into a single interface with PWA support.
chatpadChatpad AI is a self-hosted, open-source web and desktop interface for ChatGPT that stores all conversation data locally in the browser with no tracking or cookies.
ChatterMateA self-hosted AI customer support platform pairing an AI chatbot with human agents in one inbox. Works with many AI models, connects to Shopify, and spans 9 channels.
chat-uiA self-hosted, open-source chat interface for LLMs that talks to any OpenAI-compatible API (Ollama, llama.cpp, OpenRouter) with MCP tool calling and model routing.




0 comments