Ecosystem map

How the MLX ecosystem actually fits together

The main trick is to stop treating every repo like the same kind of thing. Some are foundations. Some are examples. Some help you run models. Some bridge MLX into native apps. Some are just proof that real products can be built on top.

Layer 1 Browse entries

Official foundation

The base layer: MLX itself plus the core repos Apple maintains around it.

MLX MLX C MLX ONNX ML Explore org
Layer 2 Browse entries

Docs + community

Where you learn, ask questions, and keep up with what the ecosystem is doing.

Official docs GitHub discussions Examples Guides
Layer 3 Browse entries

LLM lane

The easiest place for newcomers to get a real win: run a model locally or adapt one.

MLX LM MLX Swift LM Weights Local chat
Layer 4 Browse entries

Model discovery

The hubs where compatible weights and model listings live so you can actually test things.

mlx-community HF model browser Compatibility
Layer 5 Browse entries

Swift + native apps

For builders who want MLX inside native Mac or iPhone apps instead of only Python scripts.

MLX Swift Swift examples Native UX
Layer 6 Browse entries

Multimodal + media

The lane where MLX stops being just text and starts touching images, speech, and mixed inputs.

mlx-vlm mlx-audio MFLUX
Layer 7 Browse entries

Interop + low-level

The lower-level or bridge repos that matter once you care about portability, deployment, or integrations.

C API ONNX Bindings Interop
Layer 8 Browse entries

Apps built with MLX

Real products and interfaces built on top of MLX that make the ecosystem feel tangible.

Klee Local apps Mac-native UX

A simple mental model

Official foundation: what MLX is.

Docs + examples: how you learn it without guessing.

LLM lane: where many people get their first practical win.

Model discovery: where you find compatible weights and listings.

Swift lane: how MLX moves into native Apple apps.

Multimodal + apps: proof that MLX already reaches beyond text.

mlx-stack.txt
Foundation → examples → useful lane → stronger mental model
Docs tell you what's possible.
Examples show you what it looks like.
Curated projects show you what deserves attention.
Apps prove the stack leads somewhere real.