Google · Text
Gemma 4 E2B Instruct
Google's efficient 2B multimodal model supporting text, image, and audio — runs fully on-device via ONNX.
Modality
Text
Available on
HuggingFace
Model ID
hf:onnx-community/gemma-4-E2B-it-ONNX
Supported OS
macOSWindowsLinux
Minimum machine configuration
8 GB RAM · CPU · 128K ctx
Download size · ~5 GB
About the creator
Google and Google DeepMind build the Gemini family of multimodal models, the Imagen and Nano Banana image models, the Lyria music models, and the Veo video models.
deepmind.google ↗Samples
Examples
Sample outputs generated with Gemma 4 E2B Instruct will appear here.
Sample coming soon
Sample coming soon
Sample coming soon
More from Google
Gemini 3.1 Pro
TextCloud
Google's most capable multimodal model with deep reasoning and 1M token context.
Gemini 3 Flash
TextCloud
Fast, cost-efficient Gemini model for high-throughput text and multimodal tasks.
Gemini 2.5 Flash
TextCloud
Balanced Gemini model with strong reasoning and a 1M token context window.
Nano Banana 2
ImageCloud
Ultra-fast image generation model optimised for speed and creative output.
Lyria 2
AudioCloud
Google's music model for high-fidelity instrumental and vocal compositions.
Lyria 3
AudioCloud
Enhanced Lyria model with improved musical coherence and audio quality.