ml-benchmark

github:jiunbae/agent-skills · first seen 2026-10-31 · source · badge

ML 모델 벤치마크 및 평가 실행. "벤치마크", "모델 평가", "성능 테스트", "inference 속도" 요청 시 활성화됩니다.

Skill Health: not yet scored

Listed, not yet scored. This skill has no public model test yet. It is in the scoring queue; submitted skills go first, then the most-installed.

Version history