New chat

Gemma 4 E2B IT

Wasm heap -- Headroom -- Allocated linear memory and growth remaining below the 4 GiB Wasm limit. Not total browser RAM or free device memory.
AI runs locally in your browser

9,216-token context

Model settings

Gemma 4 E2B ITQAT mixed 2 / 4 / 8-bit

Loading speech model

Not loaded

Download
0 MB
From cache
0 MB
Load time
--
Wasm heap
--
GPU buffers
--
Memory diagnostics
Status
Load a model to collect statistics

    Cached models

    --

    Reading cache