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

Multimodal Understanding research benchmarks – Page 8 · SOTA2 Research

Multimodal Understanding

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

MMBench C (dev test)

86.05Accuracy

10

Mar 10, 2026

MMB, MME, POPE, ScienceQA (SQA), and TextVQA (VQAText)

82.8MMB Score

10

Mar 10, 2026

3.82Average Accepted Length (tau)

10

Feb 26, 2026

3.78Avg Accepted Length (tau)

10

Feb 26, 2026

0.693Score

10

Feb 26, 2026

SEED-Bench (cleaned)

88.64Overall Score

10

Feb 26, 2026

Seed-Bench (test)

InternLM-XComposer

66.9Overall Score

10

Feb 27, 2026

MMBench CN (test)

InternLM-XComposer

72.4Avg Score

10

Feb 27, 2026

Average of 10 Multimodal Benchmarks

72.15Average Score

9

Jun 17, 2026

49.71MME-R Score

9

Jun 17, 2026

MME-RealWorld in-distribution (ID) and held-out categories

84.38Diagram Accuracy

9

May 6, 2026

Multimodal Evaluation Suite Table 4

73.6ALL AVG Score

9

May 4, 2026

LLaVA-NeXT Multimodal Understanding Suite

83ChartQA Accuracy

9

May 4, 2026

MMD-Bench Hard degradation 1.0

Gemini-2.5-Flash

79.33MMBench Score

9

Apr 7, 2026

IE, MMBench, MMMU, ScienceQA Combined

100Relative Accuracy

9

Mar 17, 2026

81.6Accuracy

9

Feb 26, 2026

ScienceQA (val)

97Accuracy

9

Feb 26, 2026

MMBench en (test)

81.1Accuracy

9

Feb 26, 2026

60.3Pass@1 Score

9

Feb 26, 2026

74.94Pass@1

9

Feb 26, 2026

MMBench English (test)

85.82Score

9

Feb 26, 2026

MMBench English (dev)

86.34Overall Score

9

Feb 26, 2026

94.46Score

9

Feb 26, 2026

51.8MMD

9

Feb 26, 2026

MMBenchmark English (test)

68.8Accuracy

9

Feb 26, 2026

Page 8 of 12

Popular datasets

MMBench MMMU MME SEED-Bench MM-Vet

Follow for updates

Join SOTA2 Contact