---
title: Gemma 4 Image To Text
emoji: 🖼️
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 6.13.0
app_file: app.py
pinned: false
license: mit
short_description: Gemma 4 LiteRT-LM Image to Text
---

# Gemma 4 Image To Text

Gradio Space for concise image description with LiteRT-LM models:

- `litert-community/gemma-4-E2B-it-litert-lm`
- `litert-community/gemma-4-E4B-it-litert-lm`

Upload an image, choose a model, and receive a description only, constrained to fewer than five sentences.

The Space preloads both models at startup so switching models reuses already-open LiteRT-LM engines. On GPU hardware such as a T4, the app requests the LiteRT-LM GPU backend and falls back to CPU only if the GPU backend is unavailable or fails to initialize.