Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

SeedBench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal UnderstandingSEEDBench2 Plus
Accuracy76.86
38
Comprehensive EvaluationSeedBench (all)
Score76.8
19
Complex Scene ReasoningSEEDBench
Accuracy (All)77.03
17
Multidisciplinary KnowledgeSeedBench
Score76.4
15
Multimodal UnderstandingSeedBench image
Score76.7
12
Visual Question AnsweringSeedBench Avg
Accuracy77.96
11
Multi-discipline Video UnderstandingSeedBench video
Accuracy62.1
11
Multimodal EvaluationSeedBench (all)
Score61.82
8
VLM evaluationSEEDBench
BD-S Score12.88
6
General image understandingSEEDBench
Score72.67
4
Showing 10 of 10 rows