Google released Gemma 4 12B today. It’s a solid open-source model (Apache 2.0) that’s multimodal and runs really well on Macs with 16GB or more unified memory. Good at reasoning, coding, and agent stuff. Quick Mac-friendly info • 12B parameters, fits nicely on M2/M3/M4 Macs (especially with Q4/Q5 quant) • 256K context • Text + vision + audio support Easiest way to run it: Ollama 1. Download and install Ollama from ollama.com (the Mac app is super simple). Or use Homebrew if you prefer. 2. Open Terminal and pull the model: ollama pull gemma4:12b 3. Run it: ollama run gemma4:12b That’s it. You can start chatting right away. Mac tips: • Ollama uses Metal automatically so it runs pretty fast on Apple Silicon. • 16GB Macs handle the 12B model fine. 32GB feels even better. • Great for pairing with Continue.dev in VS Code if you code a lot. Other options if Ollama isn’t your thing: LM Studio (nice GUI), or llama.cpp for more control. Has anyone tried the image or audio features locally yet?…