big-AGI

big-AGI is a browser-based AI workspace for developers, researchers and people who need to compare answers across models. It lets you examine separate responses and combine them while keeping control over model choice and reusable context.

• Parallel answers and merging: Beam sends a prompt to several models that can't see each other's replies; Merge combines their responses using a preset or your own instructions.
• Personas with memory: Save instructions, documents, images and memory alongside a chosen model and its settings for recurring work.
• Model controls and inspection: Adjust reasoning effort, temperature and other supported settings, then inspect outgoing requests, parameters, token usage and costs with AI Inspector.
• Conversation exploration: Use split-pane chats, edit messages and branch conversations to explore alternative answers without discarding the original path.
• Voice and visual work: Use voice calls, camera or live screen input, upload files for PDF extraction, generate images through connected services, and display PlantUML or Mermaid diagrams.

The open-source foundation uses the MIT license and supports self-hosting with Docker or deployment on Vercel. A hosted service is also available, and the browser app is installable as a web app. Local model connections include Ollama, LocalAI and LM Studio; cloud connections include Anthropic, OpenAI, Google Gemini, DeepSeek and OpenRouter. Cloud providers use your own API keys and receive the prompts you send to them. The app keeps its data on your device by default.