VISTA

54.65±1.46

54.63±0.55

51.79±0.63

51.66±1.08

50.78±0.57

50.07±1.14

49.59±0.66

49.15±0.36

47.32±1.78

48.23±0.70

46.97±1.29

46.96±0.95

45.50±1.20

45.34±0.91

45.49±0.21

45.25±0.40

43.53±1.24

43.25±1.26

43.21±0.52

43.02±1.14

42.11±1.39

41.14±0.58

39.95±0.80

39.85±0.71

Claude 3.5 Sonnet (October 2024)

38.72±0.51

Claude 3.5 Sonnet (June 2024)

38.37±0.70

38.33±0.55

ChatGPT-4o-latest (November 2024)

37.99±0.48

Gemini 1.5 Pro

37.07±1.34

GPT-4o (August 2024)

34.94±0.23

34.59±1.12

Gemini 1.5 Flash 002

34.03±1.41

Pixtral Large (November 2024)

33.89±0.69

32.69±1.40

Qwen2-VL-72B-Instruct

28.56±1.37

Claude 3 Opus

27.82±0.55

26.55±0.35

Nova Pro

26.27±0.61

Pixtral 12B (September 2024)

25.97±0.74

Nova Lite

25.50±0.77

Llama 3.2 90B Vision Instruct

24.61±0.80

Llama 3.2 11B Vision-Instruct

20.47±0.15

Phi 3.5 Vision-Instruct

15.18±0.81