Gemma
Google's open model. The 12B fits a normal 16 GB laptop.
Best freeBest on your computerFreeOpen sourceRuns on your PC
- Best for
- Everyday chat on a normal laptop
- On your PC
- Some setup
- Works on
- Windows, Mac, Linux
- License
- Apache-2.0
What’s good
- Runs well on 16 GB Macs and 12 GB cards
- Understands images and short audio clips
- Speaks 35+ languages well, Apache-2.0 license
Downsides
- Less clever than Qwen3.8 27B on hard tasks
- Bigger 26B and 31B sizes need 24 GB of memory
At a glance.
- Best for
- Everyday chat on a normal laptop
- Price
- Free, open source
- Version
- Gemma 4 12BReleased 2026-06-03
- Works on
- Windows, Mac, Linux
- Open source
- Yes, Apache-2.0
- Made by
- Google DeepMind
- Website
- deepmind.google
How it ranks.
- LMArena Text (Gemma 4 31B)#77 · 14522026-10-27
- Artificial Analysis Progress Index, small open models (Gemma 4 31B)2542026-11-05
Can I run it myself?
Yes. Some setup: download a model inside an app.
Use LM Studio, Ollama, Jan, llama.cpp.
- You need
- 12 GB graphics card or Mac with 16 GB memory
- Setup
- Some setupDownload a model inside an app
- Laptop, 8 GBToo small
- PC, 16 GB+Too small
- Mac, 16 GBRuns
- Mac, 32 GBRuns
- Mac, 64 GB+Runs
- PC, 8 GB GPUToo small
- PC, 12 GB GPURuns
- PC, 16 GB GPURuns
- PC, 24 GB GPURuns
- WorkstationRuns