AI chat that stays on your terms
Your infrastructure, your rules
Everything you need. Nothing you don't.
Fully Private
Your conversations never leave your infrastructure. No telemetry, no data collection, no cloud dependencies.
Blazing Fast
Streaming responses with minimal latency. Runs on your own hardware so performance scales with you.
Self-Hosted
Deploy on your own server or locally via Ollama. You own the stack — models, data, and all.
Built for how you actually work
Everything a power user expects — thinking mode, web search, edits, and more.
Advanced Thinking Mode
Watch the model reason step-by-step with a collapsible chain-of-thought for full transparency.
Live Web Search
Give the model real-time context with toggle-on web search. Sources are automatically cited.
Image Uploads
Drag and drop images into any conversation. The model sees and analyzes them instantly.
Edit & Regenerate
Edit any past message and replay the thread. Regenerate responses until you get the perfect answer.
Pin Conversations
Keep your most important chats pinned at the top. Never lose track of ongoing work.
Export to Markdown
Download any conversation as a clean Markdown file. One click, done.
Search Conversations
Instantly surface any past chat with fuzzy search across your entire history.
Keyboard Shortcuts
Stay in the flow with power-user shortcuts for every action that matters.
Buttery Animations
Every interaction is smooth and purposeful — more polished than any chat platform you've used.
Smart Conversation Titles
Titles are auto-generated after your first message. Regenerate anytime from the sidebar to keep things organized.
Keyboard-first design
Every action has a shortcut. Stay in the flow without reaching for the mouse.
Up and running in minutes
Sign in
Create an account on your private Breeze instance.
Pick a model
Choose from any Ollama-compatible LLM running on your server.
Start chatting
Ask anything. Your data stays yours, always.