LocalOps
Back to Calculator

Gemma 3n E4B

Hot

Google's Gemma 3 nano E4B with on-demand weight loading, multimodal with text/image/audio input, 4B effective params

Model Specifications

ArchitectureVISION
Parameters8B
Familygemma
VRAM (Q4)4.0GB
On-demand architecture: weights loaded per-token. ~8GB total, ~4B effective active
#google#gemma#efficient#multimodal#edge#trendingSource

Share this Model

Send this model's specs directly to your community.

Post