ollama v0.40.0: MLX default for supported models on Apple Silicon
On Apple Silicon, models supported by the MLX runtime run on MLX by default. Adds gemma4, qwen3.8, qwen3.6, qwen3.5, decision models Nimble/tev1/clef/clef-flash, and embeddinggemma-2.
What changed
On Apple Silicon, models supported by the MLX runtime run on MLX by default; added gemma4, qwen3.8, qwen3.6, qwen3.5; decision models Nimble, tev1, clef, clef-flash; MLX gains embeddinggemma-2.
Why it matters
Developers on Apple Silicon can run supported models on MLX by default, reducing setup to use gemma4, qwen3.8, qwen3.6, qwen3.5, decision models and embeddinggemma-2 locally.
Who should care
DevelopersResearchersEnterprises
Worth trying
Maybe — if you run ollama on Apple Silicon and use the listed models