RealworldQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Real-world Understanding (RealWorldQA, MME-RW, R-Bench)
68.2RealWorld QA Score
5
RealWorldQA 90
54.51Accuracy
4
RealWorldQA
0.779Score
4
RealWorldQA
74.6Accuracy
3
RealWorldQA
268Total Inference Time (s)
2
RealWorldQA zero-shot
58.3Zero-shot Accuracy
2