GGUF
The single-file model format used by llama.cpp and many local apps, with built-in support for quantized weights.
- Open source
About Hugging Face Hub
The largest public host for open model weights, datasets and demos, where most local models are first published.
The Hugging Face Hub is where most open models are published. Each model has a page with its files, a model card describing how it was trained and what it is for, and a license. Many repositories also offer quantized versions, often in GGUF format, made for running on ordinary computers.
Local apps such as LM Studio, Ollama and llama.cpp can download models straight from it, and the site shows hints about which file sizes fit your hardware when you add your device to your profile. The same Hub hosts datasets and small demo apps.
Read the model card and license before you download. Some models are released for research only, some require you to accept terms, and some limit commercial use. Prefer files from the original publisher or a well-known uploader, and remember that a model file is code-adjacent data you are choosing to trust.
Details
Listing written by LaptopMind, checked 2026-11-04. Report a correction or request removal. Back to model hubs and formats.