Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

RealworldQA

Benchmarks

Task NameDataset NameSOTA ResultTrend
Visual Question AnsweringRealworldQA
Accuracy80.2
179
Real-world Visual Question AnsweringRealWorldQA
Accuracy77.8
140
Real-world Question AnsweringRealWorldQA
Overall Score78.7
58
Real-world Multimodal ReasoningRealWorldQA
Accuracy75.4
57
Real-world Visual UnderstandingRealWorldQA
Accuracy81.4
47
Spatial ReasoningRealWorldQA
Accuracy69.67
45
Vision-centric ReasoningRealWorldQA
Accuracy75.4
38
Visual Question AnsweringRealWorldQA (test)
Accuracy79
36
Real-world QARealworldQA
Accuracy73.1
33
Spatial UnderstandingRealWorldQA
RWQA Score66.01
30
Multimodal UnderstandingRealWorldQA
RWQA Score78
30
Real-world Visual UnderstandingRealWorldQA
Score72.29
29
General Visual UnderstandingRealWorldQA
Accuracy67.58
28
General Reasoning & UnderstandingRealWorldQA
Accuracy (RealWorldQA)72.6
21
General Visual Question AnsweringRealWorldQA
Score73.1
20
Real-world Multimodal InteractionRealWorldQA (test)
Accuracy77.8
18
Vision UnderstandingRealworldQA
Overall Score75.4
17
Visual Question AnsweringRealWorldQA (RWQA)
Score68.5
16
Real-world Multimodal InteractionRealWorldQA
RealWorldQA Score76.5
15
Visual Question AnsweringRealWorldQA 1.0 (test)
Accuracy0.6353
15
Vision-Centric UnderstandingRealworldQA
Accuracy75.4
10
Short-answer Visual Question AnsweringRealWorldQA
Accuracy65.1
9
Real-world understandingRealWorldQA
Score70.07
9
Real-world Image QARealworldQA
Score60.52
7
Real-world QARealworldQA v1.0 (test)
Score75.5
7
Showing 25 of 37 rows