Skip to content
Launchpad Library logo
All models

Open models you can run yourself

Hugging Face

Made by Hugging Face

Not one model but nearly all of them — plus the datasets, demos, courses and leaderboards.

Open Hugging FaceIts library entryCompare it with another model

What it is

Hugging Face is where open models live. Almost any model with published weights is hosted here, usually with a demo you can try in the browser before installing anything, and often with a course explaining it.

For this library it doubles as a reference desk: when a headline names a model, this is where you find out what it actually is, who released it, under what license, and how it scored.

Free accounts and downloads, with a monthly allowance of hosted inference. Paid tiers add storage, credits and team features.

Published price

  • Free100GB private storage, 100,000 inference credits a month$0
  • PRO1TB private storage, 2M inference credits$9/month
  • Team$20/user/month
  • Enterprise$50/user/month
  • Spaces hardwareFree CPU tier, GPUs by the hour

Source: Hugging Face's pricing page · Last checked 17 September 2026

Benchmarks

The published figures for Hugging Face, copied exactly as their sources give them and checked 17 September 2026. Each one names the test, who ran it and when, and says when the number comes from the maker rather than an independent tester.

No published score. Hugging Face hosts models rather than making one, so there's nothing to score. Worth knowing: its own Open LLM Leaderboard was retired in March 2025, so anyone still citing it is citing something frozen.

Where to read today's numbers

Scores move, so these are the boards that keep them current.

How it performs, in my experience

Not one model but nearly all of them — plus the datasets, demos and courses around them.

  • Free value

    5 / 5

    How much you get without paying.

  • Performance

    4 / 5

    How well it does its main job.

  • Range of uses

    5 / 5

    How many different jobs it suits.

Free and enormous — nearly every open model, plus demos and courses. These scores are my own judgment, not a measurement — the benchmark links above are the independent version.

Prompting it from this site

Can't be run from here

A hub holding thousands of models — pick one there and run it yourself; the setup steps on its page show how.

Use it for

  • Finding an open model for a specific job
  • Trying a model in the browser before installing anything

Running it yourself

Runs on your own machine

Hugging Face isn't one model — it's the hub thousands of open models are published on, and its libraries are how most local model running actually happens.

What you need first

  • Python 3.9+
  • A free Hugging Face account (needed for models with a license you must accept)
  • A GPU for anything large; small models run on a CPU

Step by step

  1. 1.Install the libraries

    pip install transformers torch huggingface_hub

  2. 2.Sign in so gated models will download

    hf auth login

  3. 3.Run a small model to prove the setup works

    python -c "from transformers import pipeline; print(pipeline('text-generation', model='Qwen/Qwen2.5-0.5B-Instruct')('Hello')[0]['generated_text'])"

  4. 4.Browse and download any other model

    hf download <org>/<model> --local-dir ./model

Check each model's license before using it for work — 'open weights' does not always mean free for commercial use.

Its own pages

Others in open models you can run yourself