brief productsopen models
Ollama 0.31 speeds Gemma 4 up to 90% on Mac
Ollama 0.31, released June 30, makes Gemma 4 up to 90% faster on Apple Silicon via multi-token prediction, on by default with no configuration needed.
sources 1 cited
1 ollama.com Faster Gemma 4 on MLX with multi-token prediction