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 10 · SOTA2 Research

Multimodal Understanding

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

81Accuracy

7

Feb 26, 2026

ScienceQA, TextVQA, GQA, VizWiz, and ImageNet

33.47Accuracy

7

Feb 26, 2026

MME standard (test)

1,803MME-P Score

7

Feb 26, 2026

ImageNet-R (test)

58.3TextVQA

7

Feb 26, 2026

COCO Caption (test)

58.3TextVQA

7

Feb 26, 2026

1,613.03Perception Total Score

6

May 28, 2026

LLaVA-Bench In-the-Wild

5.833Accuracy

6

May 21, 2026

5.6Gain Score

6

May 21, 2026

MMMU out-of-distribution

56.55Accuracy

6

Feb 26, 2026

1,487.2Score

6

Feb 26, 2026

LLaVA-NeXT Evaluation Suite

100Average Accuracy

6

Feb 26, 2026

MM-VET v1 (test)

36.4Score

6

Feb 26, 2026

LLaVA-Bench v1 (test)

74.2Score

6

Feb 26, 2026

60.15Accuracy

5

Jul 7, 2026

71.63Mean@8 Accuracy

5

May 12, 2026

MLLM Benchmark Suite

63.16GQA Score

5

Apr 24, 2026

24.66Throughput (TPS)

5

Apr 9, 2026

16.71TPS

5

Apr 9, 2026

18.22Throughput (TPS)

5

Apr 9, 2026

61.53Overall Score

5

Mar 5, 2026

ScienceQA SQA^I

90.5Accuracy

5

Feb 26, 2026

Generic Multimodal Understanding MME MMB MMStar v1.1

2,221.2MME Sum Score

5

Feb 26, 2026

Spatial Relation

1,554Perception Score

4

Jul 7, 2026

Spatial Relation

87.3Accuracy

4

Jul 7, 2026

MME-Real Perception, Reasoning, Overall

64.9Perception Score

4

Jun 24, 2026

Page 10 of 12

Popular datasets

MMBench MMMU MME SEED-Bench MM-Vet

Follow for updates

Join SOTA2 Contact