Top AI Product

ModelHub is a native macOS menu bar app that puts local-LLM management in one place. Discover models from Hugging Face, download the right local build, manage your model library, and run them through Ollama, MLX, LM Studio, or llama.cpp — without bouncing between browser tabs, terminal commands, model cards, and local folders.

## The discovery-layer pitch

Running LLMs locally on a Mac has matured — Ollama, LM Studio, MLX, and llama.cpp all work well — but the experience is fragmented. You find a model on Hugging Face, figure out which quantization fits your RAM, download it, then wire it into whichever runtime you use. ModelHub sits above all of them as a unified discovery and management layer, living in the menu bar where it’s always one click away.

## Why it matters

Local LLMs are having a moment: privacy, no API bills, offline capability. But the tooling is still developer-grade and scattered. A clean menu-bar manager that abstracts over the major runtimes is the kind of glue that moves local AI from “tinkerers only” toward “anyone with a Mac.” It doesn’t add a new runtime — it makes the existing ones livable.