enricoros/big-AGI
MULTI-MODEL AI WORKSPACE FOR EXPERTS
Open-source AI workspace for experts with 800+ model support. Allows switching models mid-chat, tuning 40+ parameters, and running Beam comparison using your own API keys.
// KEY FEATURES
- Beam and Merge for multi-model verification and de-hallucination
- 800+ models, 40+ parameters, native search and personas
- Self-host via Docker or Vercel, BYO keys
- Local-first, request inspector, no vendor lock-in
// DEEP DIVE: HOW IT WORKS
Big-AGI is a web app with local-first approach: UI runs in browser, requests go directly to provider APIs using user keys. Native protocol handles streaming, context, and parallel calls without intermediate backend.
Key feature is Beam: simultaneous multi-model execution with Merge for de-hallucination. Includes native search, personas, request inspector, 800+ models (Opus, GPT, Gemini, DeepSeek, Grok). TypeScript, deploy via Docker or Vercel.
Requires BYOK — own keys, model costs separate. Self-host: Docker or Vercel. Independent project, Pro $10.99/mo funds development. No cross-device history sync without Pro.
// USE CASES
Best for: For engineers, founders, and researchers working with multiple LLMs simultaneously
- Compare Opus, GPT, Gemini and DeepSeek responses in parallel via Beam
- Configure personas and 40+ parameters for specialized engineering tasks
- Deploy on own infrastructure via Docker with full data control
New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram