×
What is TinyML?
TinyML is a lightweight, privacy-focused chat application that runs
large language models entirely in your browser using WebGPU technology.
No servers, no cloud, no tracking.
Features
- 100% Local — Models run entirely in your browser via WebGPU
- Private — Your conversations never leave your device
- Fast — Leverages your GPU for accelerated inference
- Multiple Models — Choose from a variety of open-source LLMs
- Fine Control — Adjust temperature, tokens, top-p, top-k
- Export — Save your conversations as JSON
Technology
Built with WebLLM
from MLC-AI, enabling high-performance LLM inference directly in the browser.