GM
Gemma 4 E2B and E4B
FeaturedGoogle's edge-focused Gemma 4 models for compact multimodal local use on laptops and small devices.
Small Local Models#Gemma 4#edge model#multimodal#local AI
Gemma 4 Edge Models
Gemma 4 E2B and E4B are built for compact local deployment while still supporting multimodal input. They are useful when users need image-aware local AI without jumping directly to large workstation models.
Recommended Sizes
| Model | Package size | Context | Good for |
|---|---|---|---|
| Gemma 4 E2B | 7.2 GB | 128K | Small multimodal tasks |
| Gemma 4 E4B | 9.6 GB | 128K | Better compact local assistant |
Install Commands
ollama run gemma4:e2b
ollama run gemma4:e4b
Choose This When
- Image input matters.
- You want an edge-focused model family.
- You have at least 16 GB system memory or a GPU with enough memory headroom.