SKILL
evaluation-nemo-evaluator
Evaluates LLMs across 100+ benchmarks from 18+ harnesses (MMLU, HumanEval, GSM8K, safety, VLM) with multi-backend execution. Use when needing scalable evaluation on local Docker, Slurm HPC, or cloud platforms. NVIDIA's enterprise-grade platform with container-first architecture for reproducible benchmarking.
Target: Learning & research. Automatically collected · unverified.
Use with Codex
A guide for this combination is not available yet, and compatibility has not been verified.
Requirements
- Only the author’s Skill metadata is included. Read the full instructions to check dependencies, licensing and conditions of use.
Choose an AI tool with documented instructions, or save the resource and request a guide after the page loads.
Read the author’s documentation ↗