Text Generation
Local LLM
Choose a small language model, type a prompt, and run inference locally. Models are cached after the first download — nothing leaves your browser.
Text Generation
Run a small language model entirely in your browser. The model downloads once and is cached locally.
Model
Fast, capable small model from Hugging Face.
Prompt
Your prompt is never uploaded. Inference runs locally.