FAQ
What is Gemma 4?
Gemma 4 is Google DeepMind's latest family of open models, built from Gemini 3 research. It offers superior reasoning, multimodal capabilities, and support for 140+ languages.
How do I install Gemma 4?
The easiest way is using Ollama: simply run 'ollama run gemma4'. Alternatively, use LM Studio, Hugging Face transformers, or Docker.
What hardware do I need?
E2B and E4B models run on mobile devices. The 26B model needs ~16GB GPU memory, and the 31B model needs ~20GB GPU memory.
Is Gemma 4 free to use?
Yes, Gemma 4 is an open model with a permissive license for commercial and research use.
Can I fine-tune Gemma 4?
Yes, Gemma 4 supports fine-tuning for specific domains and use cases, allowing customization for your applications.