yichao's picture
update mj-bench
b650828
raw
history blame
3.36 kB
{
"CLIP-v1": {
"Age": 73.6,
"Gender": 75.2,
"Race": 73.1,
"Nationality": 79.1,
"Religion": 78.4,
"Avg": 75.2
},
"BLIP-v2": {
"Age": 92.2,
"Gender": 91.3,
"Race": 90.7,
"Nationality": 90.4,
"Religion": 93.1,
"Avg": 91.3
},
"PickScore-v1": {
"Age": 80.5,
"Gender": 81.2,
"Race": 81.0,
"Nationality": 81.6,
"Religion": 82.6,
"Avg": 81.2
},
"HPS-v2.1": {
"Age": 86.4,
"Gender": 87.8,
"Race": 88.5,
"Nationality": 88.0,
"Religion": 88.5,
"Avg": 87.8
},
"ImageReward": {
"Age": 85.5,
"Gender": 85.0,
"Race": 83.6,
"Nationality": 84.8,
"Religion": 89.0,
"Avg": 85.0
},
"Aesthetics": {
"Age": 91.9,
"Gender": 92.1,
"Race": 92.4,
"Nationality": 92.1,
"Religion": 92.3,
"Avg": 92.1
},
"LLaVA-1.5-7b": {
"Age": 87.4,
"Gender": 88.9,
"Race": 90.1,
"Nationality": 88.7,
"Religion": 90.7,
"Avg": 88.9
},
"LLaVA-1.5-13b": {
"Age": 87.5,
"Gender": 88.8,
"Race": 88.9,
"Nationality": 89.5,
"Religion": 90.1,
"Avg": 88.8
},
"LLaVA-NeXT-mistral-7b": {
"Age": 86.4,
"Gender": 85.8,
"Race": 85.8,
"Nationality": 84.1,
"Religion": 90.2,
"Avg": 85.8
},
"LLaVA-NeXT-vicuna-7b": {
"Age": 82.1,
"Gender": 82.8,
"Race": 82.4,
"Nationality": 82.5,
"Religion": 87.8,
"Avg": 82.8
},
"Instructblip-7b": {
"Age": 91.0,
"Gender": 91.2,
"Race": 91.1,
"Nationality": 90.4,
"Religion": 93.8,
"Avg": 91.1
},
"MiniGPT4-v2": {
"Age": 83.7,
"Gender": 83.3,
"Race": 82.8,
"Nationality": 83.4,
"Religion": 84.1,
"Avg": 83.3
},
"Prometheus-Vision-7b": {
"Age": 74.9,
"Gender": 74.3,
"Race": 73.1,
"Nationality": 74.2,
"Religion": 77.3,
"Avg": 74.3
},
"Prometheus-Vision-13b": {
"Age": 79.2,
"Gender": 76.0,
"Race": 72.7,
"Nationality": 74.1,
"Religion": 85.1,
"Avg": 76.0
},
"Qwen-VL-Chat": {
"Age": 85.9,
"Gender": 86.0,
"Race": 86.0,
"Nationality": 86.4,
"Religion": 83.8,
"Avg": 85.9
},
"Internvl-chat-v1-5": {
"Age": 86.9,
"Gender": 87.2,
"Race": 87.1,
"Nationality": 87.3,
"Religion": 88.0,
"Avg": 87.2
},
"Idefics2-8b": {
"Age": 77.0,
"Gender": 79.7,
"Race": 81.3,
"Nationality": 82.0,
"Religion": 74.4,
"Avg": 79.8
},
"GPT-4-vision": {
"Age": 93.0,
"Gender": 93.2,
"Race": 92.2,
"Nationality": 93.4,
"Religion": 96.4,
"Avg": 93.2
},
"GPT-4o": {
"Age": 91.8,
"Gender": 92.9,
"Race": 93.1,
"Nationality": 93.3,
"Religion": 94.4,
"Avg": 92.9
},
"Gemini Ultra": {
"Age": 86.6,
"Gender": 89.0,
"Race": 90.8,
"Nationality": 90.0,
"Religion": 86.2,
"Avg": 89.0
},
"Claude 3 Opus": {
"Age": 83.2,
"Gender": 85.2,
"Race": 86.5,
"Nationality": 85.8,
"Religion": 84.8,
"Avg": 85.2
}
}