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
Adversarial Attack	FLIR ADAS 1.3 (test)	adversarial safety securitycomputer vision	35	Feb 26, 2026
Entity Disambiguation	ZELDA Benchmark (test)	nlp	35	Feb 26, 2026
Mathematical Reasoning	SVAMP	llm chat instructionnlp	35	Feb 26, 2026
Image-to-Sentence Retrieval	MS-COCO 5K full (test)	computer visionnlp	35	Feb 26, 2026
Object Detection	COCO-O	computer vision	35	Mar 31, 2026