LiteRT-LM

by Infinitude
Unloaded

Gemma 4 E2B IT Web (Multimodal & Text)

LiteRT-LM Gemma 4 Chat

Fast, local-first AI inference running directly in your browser via WebGPU. Supports text, code files, multimodal images, and Whisper voice inputs.

Designed & Built byInfinitude

Model Not Loaded

Download or select local model binary

Suggested Prompts

Local WebGPU Multimodal & Whisper Voice Inference ยท Zero Data Transmitted