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
Mathematical Reasoning	MATH-500	llm chat instructionnlp	33	Jul 9, 2026
High-Resolution Image Perception	V* Bench	computer vision	33	Jun 24, 2026
Adversarial Detection	ImageNet BLIP-2	adversarial safety security	33	May 25, 2026
Audio Watermark Detection	Stable Audio generative edits (test)	speech audio	33	May 25, 2026
Generalized Category Discovery	SSB Average	computer vision	33	May 22, 2026