Share your thoughts, 1 month free Claude Pro on usSee more

Aggregate Multimodal Evaluation on Average 8 Benchmarks

74.4Average Score

Qwen-VL-Max-0809

Updated 1mo ago

Evaluation Results

Method	Links
Qwen-VL-Max-0809 2024.10		74.4
GPT-4o-20240806 2024.10		71.5
InternVL2-Llama3-76B 2024.10		71
Claude3.5-Sonnet 2024.10		67.9
Gemini-1.5-Pro 2024.10		64.4
Qwen2-VL-2B 2024.10		57.2
GPT-4v (1106, detail-high) 2024.10		56.4
Gemini-1.0-Pro 2024.10		56.1
H2OVL-Mississippi-2B 2024.10		54.4
InternVL2-2B 2024.10		53.9
Phi-3-Vision 2024.10		53.6
Claude3-Sonnet 2024.10		53.5
MiniMonkey 2024.10		52.7
Qwen-VL-Plus 2024.10		52.2
InternVL2-1B 2024.10		48.3
MiniCPM-V-2 2024.10		47.9
PaliGemma-3B-mix-448 2024.10		46.5
H2OVL-Mississippi-0.8B 2024.10		43.5
DeepSeek-VL-1.3B 2024.10		39.6