Agent Skills: nemo-evaluator-sdk

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.

UncategorizedID: davila7/claude-code-templates/nemo-evaluator-sdk

Author

davila7

https://github.com/davila7 View all skills

Repository

davila7/claude-code-templates

davila7License: MIT

19,6461,834

Install this agent skill to your local

pnpm dlx add-skill https://github.com/davila7/claude-code-templates/nemo-evaluator-sdk

Skill Files

Browse the full folder contents for nemo-evaluator-sdk.

Download Skill

Loading file tree…

Select a file to preview its contents.