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
Multi-attribute steering	ToneBank	llm chat instruction	42	Jun 23, 2026
Vision-language alignment	CINIC-10	multimodalcomputer vision+1	42	Jun 23, 2026
Vision-language alignment	CIFAR-10	multimodalcomputer vision+1	42	Jun 23, 2026
Generalized Category Discovery	CIFAR-10	computer vision	42	Jul 7, 2026
Spacecraft Trajectory Forecasting	NASA CDDIS RTN Frame Extrapolation Horizons 5760M	tabular time series