List your company

ResearchFormerly WizWand

Multimodal Understanding

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

High-IC tasks MMB, POPE, MME, SEED, GQA, TextVQA LLaVA-NeXT-7B

95.7Performance Ratio

18

Mar 4, 2026

LLaVA High-IC tasks (MMB, POPE, MME, SEED, GQA) 1.5-7B

94.7Performance Ratio

18

Mar 4, 2026

73.8ShopFront Score

18

Feb 26, 2026

Multimodal Evaluation Suite MMB, MME, SQA, VQA^T, MMB^C, MMVet, MMstar, AI2D

64.6MMB Score

17

Feb 26, 2026

Multimodal Evaluation Suite GQA, MMBench, MMBench-CN, MME, POPE, ScienceQA, TextVQA (various)

Vanilla (Qwen2-VL-7B)

62.2GQA Accuracy

16

Jun 24, 2026

80.56Accuracy (Multi-Choice)

16

May 19, 2026

Bagel+LatentUMM

1,696.1MME Score

16

May 19, 2026

Aggregate Non-MME Suite

86.62Average Accuracy

16

May 19, 2026

HallusionBench, MME, AI2D, RealWorldQA, ScienceQA, POPE, MMBench (EN/CN), CCBench, Visual Spatial Reasoning, Visual7W (test)

58.09HallusionBench Score

16

May 19, 2026

87.64MBen Score

16

May 19, 2026

2,388MME Score

16

May 13, 2026

2,433.9Score

16

Apr 2, 2026

SEED-Bench (val)

58.8Accuracy

16

May 26, 2026

80.8Accuracy

16

Mar 27, 2026

VQAv2, GQA, VQAText, POPE, MME

81.2VQAv2 Accuracy

16

Feb 26, 2026

200Existence Score

16

Apr 8, 2026

6,023Latency

16

Apr 7, 2026

AI2D, POPE, TextVQA, OKVQA, VizWiz, COCO Caption, NoCaps, and RealWorldQA Average

99.86Average Score

15

Jun 15, 2026

67.2MMVet Score

15

Apr 28, 2026

Multimodal Understanding Benchmarks (CQA, DVQA, TVQA, IVQA, OCRB, MMB, MMMU, MVista, AI2D, MMS, MMV) official (test)

85.8MMB

15

Feb 26, 2026

MME-RW-en (test)

71.4Overall Score

15

Feb 26, 2026

Gemini 2.5 Flash

79.9Average Score

15

May 1, 2026

SEED Benchmark (test)

73.8Avg Score (All)

15

Apr 9, 2026

73.5Image Accuracy

15

Feb 26, 2026

SEED-Bench Image Part

InternLM-XComposer2

75.9Accuracy

15

Feb 26, 2026