huggingface-community-evals

Official

Local HF model evals with flexible backends

AuthorLAF-US
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This skill enables teams to run local, configurable evaluations of Hugging Face Hub models using inspect-ai and lighteval with multiple backend options. It centralizes local evaluation workflows for quick testing and backend comparisons without relying on HF Jobs orchestration.

Core Features & Use Cases

  • Local evaluation of Hugging Face Hub models using inspect-ai and lighteval with backends such as vLLM, Transformers, and accelerate.
  • Script-driven deterministic runs with separate scripts for smoke tests, backend selection, and task management.
  • Safe hand-off boundaries for remote job orchestration and publishing workflows to dedicated automation skills.

Quick Start

Run a local Hugging Face Hub model evaluation using the provided inspect-ai or lighteval scripts.

Dependency Matrix

Required Modules

inspect-ai>=0.3.0inspect-evalsopenaivllm>=0.4.0torch>=2.0.0transformers>=4.40.0lighteval[accelerate,vllm]>=0.6.0accelerate>=0.30.0

Components

scripts

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: huggingface-community-evals
Download link: https://github.com/LAF-US/IDAHO-VAULT/archive/main.zip#huggingface-community-evals

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.