Skip to content
mlx.app

MLX-VLM

The go-to package for running vision-language models with MLX.

When to use it

Use this when you need Qwen-VL, LLaVA or Idefics locally and want mlx-lm-style CLI ergonomics for image plus text prompts.

Alternatives in Vision

MobileCLIP

Apple's on-device-sized CLIP for fast image-text matching.

OpenCLIP

The reproducible, community-trained CLIP zoo everyone benchmarks against.

Ferret

Apple's research VLM for referring and grounding inside images.