Ollama v0.31.1 Speeds Up Gemma 4 on Apple Silicon by Nearly 90%

Core summary

One sentence to understand this update

Ollama v0.31.1 was released, boosting Gemma 4 model performance on Apple Silicon by nearly 90% faster token generation, leveraging multi-token prediction.

Impact & opportunity

What this could mean

This update significantly enhances Gemma 4's efficiency for developers performing local AI model development and inference on Apple Silicon devices, accelerating testing and iteration cycles.