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

Multimodal Mathematical Reasoning on OlympiadBench

68Accuracy

o1

Updated 8d ago

Evaluation Results

Method	Links
o1 2026.03		68
o1 2026.03		68
Gemini2-flash 2026.03		51
Gemini-2-flash 2026.03		51
Claude3.7-Sonnet 2026.03		48.9
Claude3.7-Sonnet 2026.03		48.9
MM-Eureka-32B-R-TAP 2026.03		41.2
Qwen-2.5-VL-72B 2026.03		40.4
Qwen-2.5-VL-72B 2026.03		40.4
MM-Eureka-32B 2026.03		35.9
Claude3.7-Sonnet 2025.09		35.2
GPT-4o 2026.03		35
GPT-4o 2026.03		35
QVQ-72B-Preview 2026.03		33.2
QVQ-72B-Preview 2026.03		33.2
InternVL2.5-VL-38B 2026.03		32
InternVL2.5-VL-38B 2026.03		32
InternVL2.5-VL-78B 2026.03		31.1
InternVL2.5-VL-78B 2026.03		31.1
Qwen-2.5-VL-32B 2026.03		30
Qwen-2.5-VL-32B 2026.03		30
GPT-4.1-20250414 2025.09		29.4
MM-Eureka-7B-R-TAP 2026.03		27.5
GPT-4o 2025.09		25.9
InternVL2.5-38B-MPO 2026.03		25.6
InternVL2.5-38B-MPO 2026.03		25.6
DIVA-GRPO-7B 2026.03		23.1
Qwen2.5VL-72B x Qwen3-32B 2025.09		22.4
R1-ShareVL-7B 2026.03		21.3
Qwen-2.5-VL-7B 2026.03		20.2
Qwen-2.5-VL-7B 2026.03		20.2
QVQ-72B-Preview 2025.09		20.2
Adora-7B 2026.03		20.1
OpenVLThinker-7B 2026.03		20.1
MM-Eureka-7B 2026.03		20.1
ADORA-7B 2026.03		20.1
OpenVLThinker-7B 2026.03		20.1
MM-Eureka-7B 2026.03		20.1
OpenVLThinker-7B 2025.09		20.1
Qwen2.5-VL-72B 2025.09		19.9
R1-Onevision 2025.09		17.8
Qwen2.5VL-7B x Qwen3-32B 2025.09		17.4
R1-Onevision-7B 2026.03		17.3
R1-Onevision-7B 2026.03		17.3
SFT-7B 2026.03		16.2
InternVL3.5-8B 2025.09		16.1
Keye-VL-8B 2025.09		15.4
Qwen2.5VL-7B x Qwen3-4B 2025.09		14.7
SAIL-VL2-8B 2025.09		14.1
InternVL2.5-VL-8B 2026.03		12.3
InternVL2.5-VL-8B 2026.03		12.3
Qwen2.5-VL-7B 2025.09		8.6
InternVL2.5-8B-MPO 2026.03		7.8
InternVL2.5-8B-MPO 2026.03		7.8
VLM-R1-3B-Math 2025.09		7.3
Qwen2.5-VL-3B 2025.09		7