List your company

ResearchFormerly WizWand

Multimodal Visual Question Answering

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

88.6Score

18

Jun 1, 2026

LLaVA Evaluation Suite (GQA, MME, POPE, SQA-Img, VizWiz, VQAv2, MMB-En) 1.5

61.9GQA

16

May 22, 2026

VLMEvalKit Image Benchmarks

46.8HallBench Accuracy

13

Apr 10, 2026

32.62Accuracy

6

Mar 9, 2026

27.18Accuracy

6

Mar 9, 2026

GQA, MMB, POPE, and MME Average Retention LLaVA-v1.5-7B (test)

Vanilla (LLaVA)

100Accuracy

4

Jun 16, 2026