Benchmarks

Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.

309,131 indexed

Task	Dataset	Domain	Results	Last Update
Image Classification	CIFAR-10 Dir(0.5) (test)	computer vision	36	Jun 23, 2026
OOD Detection	OpenOOD CIFAR10 Near-OOD	computer visiontabular time series	36	May 27, 2026
LLM-as-a-Judge	RewardBench 1.0 (test)	llm chat instruction	36	Feb 26, 2026
LLM-as-a-Judge Evaluation Consistency	RewardBench	llm chat instructionadversarial safety security	36	Feb 26, 2026
Medical Knowledge Evaluation