Gemma · 26B · Desktop / import
Gemma 4 26B-A4B Instruct on a phone
MoE Gemma 4: 26B total, ~4B active. Big-model taste at 8B-class compute — GPU or 32 GB+ Mac, then Connect.
- Parameters
- 26B
- Typical Q4
- ~8–12 GB (Q4 MoE)
- RAM to budget
- ~14 GB
- Engines
- GGUF
Get it running
google/gemma-4-26B-A4B-it. Official QAT GGUF. Efficiency king of the Gemma 4 stack. Hugging Face: google/gemma-4-26B-A4B-it.
Load in LM Studio or Ollama, then chat from the phone.
Get LM Mini and load a model that actually fits — on-device, or via Connect to a PC.
Phones that can run it
- None in this guide — use a Mac or Connect.
Tight (close other apps)
- None in this guide — use a Mac or Connect.