Holistic Evaluation of Language Models (HELM)

Beyond the Imitation Game collaborative benchmark for measuring and extrapolating the capabilities of language models - google/BIG-bench

Measuring Massive Multitask Language Understanding | ICLR 2021 - hendrycks/test

LiveBench AI模型实时评测平台，追踪和比较各模型性能变化

SWE-bench Leaderboards