List your company

ResearchFormerly WizWand

MMBench

Benchmarks

TaskDatasetSOTA resultTrendResults

Multi-modal Benchmark Evaluation

MMBench v1.0 (dev)

69.2MMB^dev Score

15

MMBench en (dev)

90.3Accuracy

15

Visual Perception

MMBench (T)

79.1Accuracy (MMBench T)

15

Multi-modal Benchmark

MMBench (dev)

73.5Accuracy

15

Visual Question Answering

MMBench

89.97Class Accuracy

14

Multimodal Understanding

MMBench

4.44ECE

14

Multimodal Understanding

MMBench OOD

99.84Accuracy

14

Malicious Prompt Detection

MMBench OOD

0.16FPR

14

Multi-modal Vision-Language Understanding

MMBench CN

64.7Overall Score

14

Multimodal Understanding

MMBench v1.0 (test)

67.7Accuracy

14

Multimodal Model Evaluation

MMBench EN

83.9Accuracy

13

General Multimodal Benchmarking

MMBench Chinese (val)

82.4Accuracy

13

General Multimodal Benchmarking

MMBench English (val)

80.2Accuracy

13

Multi-modal Perception

MMBench (dev)

69.2MMB Score (dev)

13

Multimodal Evaluation

MMBench en

54MMBench Score

13

Multimodal Question Answering

MMBench

83.6MMBench Score

13

MMBench EN (test)

32MMBenchEN Score

13

Multimodal Question Answering

MMBench-CN (dev)

73Accuracy

13

Multimodal Question Answering

MMBench (dev)

77.3Accuracy

13

Multimodal Understanding

MMBench en cn

62.6MMBench English Score

13

General-purpose multiple-choice evaluation

MMBench EN V11 (dev)

84.3Pass@1

13

General Visual Question Answering

MMBench 2024c (dev)

83.3Accuracy

13

Image Understanding

MMBench en 2024 (dev)

85Score

12

Image Understanding

MMBench EN 1.0

86Score

12

Video Understanding

MMBench-Video

1.46TR Score

12