Big-AGI is an open-source multi-model AI workspace with Beam, multi-model chats, and local-first deployment options. It emphasizes AI personas, multi-model integrations, and on-prem or cloud deployment.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
Big-AGI is an open-source AI suite focused on multi-model AI workspace capabilities, including AI personas, AGI functions, and multi-model chats via Beam. It supports text-to-image, voice, response streaming, code highlighting and execution, PDF import, and developer presets. It can be deployed on-premises or in the cloud.
How it works
The project aggregates multiple AI models and services through Beam for multi-model management, with features like persistent sessions via Anthropic Containers in recent releases. It supports local-first deployment and API-key based usage, enabling cross-model workflows and inspection through AI Inspector.
Getting started
- Self-host and developers (full control): "Develop locally or self-host with Docker on your own infrastructure – guide"
- Or fork & run on Vercel: deployment link;
- Installation guidance is available in the docs: "installation.md" under docs.
Commands and setup details are provided in the README and release notes, including references to installation pages and environment variables.
Recent releases
- v2.0.5 (2026-05-13): Big-AGI 2.0.5 with Opus 4.7 1M, GPT-5.5, DeepSeek V4, Anthropic Containers/Skills, resumable Deep Research, MiniMax & Arcee AI
- v2.0.4 (2026-03-25): Hyper Params with precise parameter ranges for reasoning effort, verbosity, and temperature across providers
- v2.0.3 (2026-02-02): Red Carpet with improved parameter configuration, Gemini 3 Flash, GPT 5.2, and cloud/service integrations
- v2.0.2 (2025-12-01): Speex – multi-vendor speech synthesis, Opus 4.5, Gemini 3 Pro, Nano Banana Pro, Grok 4.1, GPT-5.1, Kimi K2
- v2.0.1 (2025-11-24): Heavy Critters
Traction
- Stars: 7075
Behind the repo
- Repository is by enricoros; other details such as license and creation date are listed in the FACTS block.
Caveats
- License: MIT
- Created: 2023-03-19
- Last push: 2026-08-03
- Open issues: 278
- Language: TypeScript
- Forks: 1594
- Topics include agi, ai-agents, ai-suite, ai-workspace, multi-modal, multi-model, and various vendor integrations.