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
Variable Tracking	RULER-VT	llm chat instruction	33	Mar 12, 2026
Key-value Retrieval	LITM (Lost in the Middle)	retrieval search rag	33	Mar 12, 2026
Graph Anomaly Detection	Yelp	graph	33	May 13, 2026
Text Retrieval	MSCOCO	retrieval search rag	33	Mar 6, 2026
Image Denoising	BSD68	computer vision	33	May 13, 2026
Instruction Following	Instruction-Following (val)