Holistic Evaluation of Language Models (HELM)Stanford CRFM
Open sourceOpen source framework for transparent, holistic evaluation of language models.
AI Evaluation & Benchmarking Frameworks
Ranking methodology
In development — no version effective yet.
Decision archetype: ai · Scores are category-relative (P-01) and commercial relationships never affect rank (P-04).
Tools (3)
InspectUK AI Security Institute
Open sourceOpen source framework for evaluating and stress-testing large language models.
LM Evaluation HarnessEleutherAI
Open sourceOpen source framework for benchmarking language models across standardized tasks.