Open-source Gemini/ChatGPT-like interface for local browser-based open-source models.
What is Chatty?
Chatty is an open-source, feature-rich interface similar to Gemini/ChatGPT. Its purpose is to enable the local execution of open-source models such as Gemma, Mistral, and Llama3 within browsers via WebGPU. It eliminates the need for server-side processing, ensuring that user data remains on the user’s personal computer.
How to use Chatty?
To use Chatty, simply select a model (such as Gemma, Mistral, or Llama3) and initiate a new conversation. Enter your query in the chat interface, and you will receive the response directly within your browser – without any server-side processing involved.
The core functions of Chatty
Executing open-source models in a local browser
Efficient processing using WebGPU
Privacy-conscious data processing (no server-side processing involved)
An interface similar to that of Gemini/ChatGPT