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

Diagram Understanding on AI2D 1.0 (test)

96.3Accuracy

Molmo-72B

Updated 1mo ago

Evaluation Results

Method	Links
Molmo-72B 2024.12		96.3
InternVL2.5-78B 2024.12		95.7
InternVL2.5-38B 2024.12		95.1
InternVL2-Llama3-76B 2024.12		94.8
Claude-3.5-Sonnet 2024.12		94.7
InternVL2-40B 2024.12		94.5
InternVL2.5-26B 2024.12		94.4
Gemini-1.5-Pro 2024.12		94.4
GPT-4o-20240513 2024.12		94.2
NVLM-D-72B 2024.12		94.2
Molmo-7B-D 2024.12		93.2
InternVL2.5-8B 2024.12		92.8
InternVL2-26B 2024.12		92.5
Qwen2-VL-7B 2024.12		92.1
InternVL2-8B 2024.12		91.7
InternVL2.5-4B 2024.12		90.5
InternVL-Chat-V1.5 2024.12		89.8
GPT-4V 2024.12		89.4
InternVL2.5-78B 2024.12		89.1
Claude-3-Opus 2024.12		88.1
Qwen2-VL-72B 2024.12		88.1
InternVL2-4B 2024.12		87.8
Phi-3.5-Vision-4B 2024.12		87.6
InternVL2.5-38B 2024.12		87.6
InternVL2-Llama3-76B 2024.12		87.6
InternVL2-40B 2024.12		86.6
InternVL2.5-26B 2024.12		86.4
LLaVA-OneVision-72B 2024.12		85.6
NVLM-D-72B 2024.12		85.2
Qwen2-VL-2B 2024.12		84.6
GPT-4o-20240513 2024.12		84.6
InternVL2.5-8B 2024.12		84.5
InternVL2-26B 2024.12		84.5
Ovis1.6-Gemma2-9B 2024.12		84.4
InternVL2-8B 2024.12		83.8
InternVL2.5-2B 2024.12		83.5
Qwen2-VL-7B 2024.12		83
InternVL2-2B 2024.12		82.3
MiniCPM-V2.6 2024.12		82.1
InternVL2.5-4B 2024.12		81.4
Claude-3.5-Sonnet 2024.12		81.2
InternVL-Chat-V1.5 2024.12		80.7
Cambrian-34B 2024.12		79.5
Gemini-1.5-Pro 2024.12		79.1
InternVL2-4B 2024.12		78.9
GPT-4V 2024.12		78.2
InternVL2.5-1B 2024.12		77.8
Phi-3.5-Vision-4B 2024.12		77.8
Aquila-VL-2B 2024.12		75
InternVL2.5-2B 2024.12		74.9
Qwen2-VL-2B 2024.12		74.7
InternVL2-2B 2024.12		74.1
Claude-3-Opus 2024.12		70.6
InternVL2-1B 2024.12		70.5
VILA-1.5-40B 2024.12		69.9
InternVL2.5-1B 2024.12		69.3
InternVL2-1B 2024.12		64.1
LLaVA-OneVision-0.5B 2024.12		57.1