List your company

ResearchFormerly WizWand

Multimodal Understanding and Reasoning

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

MMMU, SEED, OCRBench, VizWiz, ScienceQA, TextVQA (test/val)

61.9MMMU Score

42

Feb 26, 2026

78.7MME Score

26

Feb 26, 2026

MMBench Chinese (dev)

72.8Accuracy

22

Feb 26, 2026

Multimodal Evaluation Suite (MMVet, MMBench_EN, SEED-Bench, LLaVABench, POPE, MME-P, MMVP, MMStar) (Random Sampling Splits of CC12M)

30.1MMVet Score

13

Apr 28, 2026

Image Benchmarks HallBench, MME, TextVQA, ChartQA, AI2D, RealWorldQA, CCBench, OCRVQA, SQA-IMG, POPE

46.5HallBench Score

13

Apr 10, 2026

Multimodal Evaluation Suite (VQAv2, GQA, SQAImg, VQAText, POPE, MME, MMBEN, MMBCN, MMVet)

81.3VQAv2 Accuracy

10

Jun 18, 2026

59Accuracy

7

May 27, 2026

MLLM Evaluation Suite (MME, MMB, VizWiz, POPE, GQA, RQA, VQAT, SQA) standard (test val)

2,375MME Score

5

May 29, 2026

44.6MM-Vet Score

4

Feb 26, 2026

44.7Accuracy

3

May 27, 2026

MMBench Tibetan

67.78Overall Score

2

May 27, 2026