Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

General Multimodal Evaluation research benchmarks · SOTA2 Research

General Multimodal Evaluation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

69.7Overall Score

16

May 28, 2026

69.1Accuracy

14

May 28, 2026

Macro-average of HallusionBench, AMBER, CRPE, R-Bench, and BLINK

63.35Overall Score

13

Jun 1, 2026

LLaVA-Bench Wild

92.8Relative Score

12

Feb 26, 2026

MMBench CN (dev)

LLaVA-1.5 + FG-CLIP 2

60.5Accuracy

4

May 26, 2026

MMBench EN (dev)

LLaVA-1.5 + FG-CLIP 2

67.6Accuracy

4

May 26, 2026

Popular datasets

MMBench MMMU LLaVA-Bench MMVet Macro-average of HallusionBench, AMBER, CRPE, R-Bench, and BLINK

Follow for updates

Join SOTA2 Contact