GenEval
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GenEval
0.733Overall Score
12
GenEval
100Single Object Accuracy
11
GenEval
87Geneval Score
10
GenEval 17 (test)
1.59Latency (s)
10
GenEval Text Rendering and Human Preference Evaluation Suite (test)
0.95GenEval Score
10
GenEval++
90Color Accuracy
10
GenEval 512px
99Compositional Score: Single Object
9
GenEval++ (test)
90Color Accuracy
9
GenEval (test)
90.3Mean Score
9
GenEval SD V1.5
57Overall Score
9
GenEval (test)
0.14Latency (s)
8
GenEval
0.7DINO
8
GenEval FLUX.1 (dev)
69.88Overall Score
8
GenEval
78Overall Score
8
GenEval zero-shot
0.55GenEval Score
8
GenEval 11 (test)
0.275Overall Score
8
GenEval original (test)
99Single Object Accuracy
8
GenEval
89Aesthetic Quality
7
GenEval
94YOLOv9 Score
7
GenEval
92Accuracy (2 objects)
7
GenEval
75.58MUSIQ Score
7
GenEval
100Accuracy (Single Object)
7
GenEval out-of-domain
69.62Semantic Consistency
7
GenEval short prompts original
0.48GenEval Score
6
GenEval
0.07KV Size (GB)
6