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
Math	GSM8K	llm chat instruction	47	Apr 15, 2026
Text-to-Image Retrieval	RSITMD	multimodalretrieval search rag	47	Apr 23, 2026
Image-to-text Retrieval	RSITMD	multimodalretrieval search rag	47	Apr 23, 2026
Reasoning Segmentation	EarthReason (val)	computer vision	47	May 25, 2026
Visual Rating (Image Aesthetic Assessment)	FLICKR-AES	computer vision	47	Jun 5, 2026
Vehicle Re-Identification