A workflow for selecting GGUF model files and running or serving them locally with llama.cpp.
Editorial review checked 2026-10-29 against this instruction snapshot. Supporting scripts and runtime behavior remain outside this editorial review.
This skill connects model discovery with the practical details of selecting an exact GGUF file and quantization. It distinguishes using an existing compatible artifact from converting model weights, and includes local command-line and server-oriented paths. The result should be a reproducible model choice tied to the machine and workload, rather than an unspecified recommendation to download a large model.
Preparing a local inference setup, comparing model sizes against available memory, or diagnosing a repository whose file naming differs from standard launcher expectations. Describe the hardware, available RAM or VRAM and the tasks that matter to the user.
Quantization changes memory use and may change output quality. Local operation still requires model downloads and appropriate model licenses; it does not mean every listed model fits every device. Gated repositories and a network-accessible local server introduce additional access considerations.
Unknown; inspect source requirements and supporting scripts.
Inspect the pinned source snapshot and its supporting files first. The command selects this skill through the open-source skills CLI. Node.js and agent-specific setup may be required. Installation can download code. Catalog ingestion never executes skill instructions or install scripts.
npx skills add https://github.com/huggingface/skills/tree/25910b3d0ba31a2ba102f901cdc9e008c278a69c/skills/huggingface-local-models --skill huggingface-local-modelsReview pinned skill folder ↗ · CLI documentation ↗
After source review, use npx skills check and npx skills update to inspect and apply upstream updates. Re-check changed instructions before running them.
Static review only. This is not a security audit or a guarantee of safety.
SKILL.md text only; scripts, dependencies and runtime behavior not audited.
Allowed tools declaration: Unknown. Runtime permissions and network behavior: Unknown.
87/100 · Metadata coverage 100% · Model 1.0
Scores summarize observable discovery signals. Popularity and repository activity apply to the whole repository. Unknown runtime safety earns no certification.
Choose a compatible model family and exact file, verify its requirements, run a small representative task and only then configure ongoing local serving. Preserve the selected repository, file and launch settings.
Measure memory use, startup behavior and useful task quality on the target hardware. Check the server's binding address and confirm that any claimed offline operation works without a new download.
Related by category; compare scope and dependencies before choosing an alternative.
Explore advanced evaluation as a ai & machine learning workflow from sickn33/agentic-awesome-skills. Compare its source, maintenance and review signals before choosing it for your agent.
Explore aegisops ai as a ai & machine learning workflow from sickn33/agentic-awesome-skills. Compare its source, maintenance and review signals before choosing it for your agent.
Explore agentic actions auditor as a ai & machine learning workflow from sickn33/agentic-awesome-skills. Compare its source, maintenance and review signals before choosing it for your agent.
Explore hf cloud aws context discovery as a cloud & devops workflow from huggingface/skills. Compare its source, maintenance and review signals before choosing it for your agent.
Explore hf cloud python env setup as a development workflow from huggingface/skills. Compare its source, maintenance and review signals before choosing it for your agent.
Explore hf cloud sagemaker deployment planner as a cloud & devops workflow from huggingface/skills. Compare its source, maintenance and review signals before choosing it for your agent.
Editorial combinations of complementary roles; review each source independently.
Inspect a bounded dataset, select an exact local model artifact and build an understandable demonstration interface.