What would using
Happy Chat be like?

Lightning-fast AI conversations with streaming responses, real-time thinking display, and dynamic model routing.

Streaming
Watch responses appear token by token
🧠
Thinking
See the AI reason through problems live
🔀
Routing
Automatically picks the best model
Enter — Send Ctrl+Shift+N — New Chat Ctrl+, — Settings
0 tok/s
Enter to send · Shift+Enter for newline · Markdown supported
Settings
Think Mode (reasoning)
Store Chat History
Request a Model

Request an Ollama model to be added to the inference cluster. An admin will review your request.