TinyServe
WebGPU · ONNX Runtime Web
Watch a language model think, live in your browser.
- Your prompt
- Your GPU
- Next word + odds
WebGPU · ONNX Runtime Web
Watch a language model think, live in your browser.
Watch a language model think, live in your browser.
Turn them off in Settings.
Built with Llama. Llama 3.2 is licensed under the Llama 3.2 Community License, Copyright © Meta Platforms, Inc. All Rights Reserved. LFM2.5 is licensed under the LFM Open License v1.0 by Liquid AI. License texts · Llama use policy
Want a similar tool built for your company? Write to us.
Runs in your browser: the model writes its answers on your GPU, not on a server. With WebGPU, opening this page already asks huggingface.co for the file sizes, so that site sees your IP address.
One thing does carry your prompt: the address bar. A link you copy includes it, and opening or reloading that address sends it to the server that hosts this page.