What you can do
- Choose GGUF models and quantizations that fit local hardware with llama.cpp.
Before you start
- llama.cpp, sufficient RAM/disk space, and network access for model downloads.
Try your first task
After installing, send this example to your AI tool.
Use huggingface-local-models. Compare suitable GGUF options for my 16 GB machine and show how to run the smallest viable model.
What to look for
Hardware-aware model choices and a command using a verified model filename.
Keep in mind
- Model licenses, gated access, memory needs, and speed vary independently of this Skill.
Open source, traceable origins
Source, license, and package contents have been reviewed. End-to-end tasks have not been verified across agents; examples describe expected results.
Source reviewed:
FAQ
Agent Skills FAQ
Practical answers for choosing, installing, and using your first skill.
How do I install and start using Run local language models?
Download the complete ZIP and keep the huggingface-local-models folder and supporting files together. For a project installation, use .claude/skills/huggingface-local-models/ in Claude Code or .agents/skills/huggingface-local-models/ in Codex, with SKILL.md directly inside it. Complete this page's prerequisites, confirm discovery, and start with the example in the first-task card.
What do I need to use Run local language models?
llama.cpp, sufficient RAM/disk space, and network access for model downloads.
Can I use these Skills with Chinese requests?
You can ask an agent that supports Chinese to work with Chinese materials and produce Chinese output. Specify the desired language, audience, and terminology in your request. Results depend on the agent and skill, so check the output against your task requirements.
Who created this skill and which license is included?
The original author is Hugging Face, from huggingface/skills, with the Apache-2.0 license included. AILesson preserves the files and source version, and GitHub Releases provides the download. Use the source card to inspect the pinned version and original files.
Have all Skills been tested, and what should I review?
The current collection has source, license, and package-content checks. It does not claim that every skill has completed real tasks in every agent. Review the files and prerequisites before enabling a skill, and check facts, citations, generated code, and proposed actions in its output. The examples describe expected results rather than verified demonstrations.
