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

Visual Mathematical Reasoning on MathVista (score)

81.9Score

GPT-5-Thinking

Updated 1mo ago

Evaluation Results

Method	Links
GPT-5-Thinking 2025.09		81.9
Gemini-2.5-Pro 2025.09		80.9
CGC-8B 2026.04		78.2
VAPO-Thinker-7B 2025.09		75.6
Qwen3-VL-8B 2026.04		75.3
Vision-R1-7B 2025.09		73.5
MM-Eureka-7B 2026.04		72
ThinkLite-VL-7B 2026.04		71.89
NoisyRollout-7B 2026.04		71.6
CGC-7B 2026.04		70.9
VLAA-Thinker-7B 2026.04		70.8
InternVL3-8B 2025.10		70.5
QvQ-72B-Preview 2025.10		70.3
CHARTOOL-7B 2026.04		70.1
DRIFT 2025.10		69.9
X-REASONER 2025.10		69
SFT Baseline 2025.10		68.7
InternVL2.5-8B 2025.09		68.2
Qwen2.5-VL-7B 2026.04		68.2
VLAA-Thinker-7B 2025.09		68
HeadLens 2026.03		67.9
MiCo-7B 2026.04		67.9
Qwen2.5-VL-7B 2025.10		67.9
InternVL3-8B 2026.04		67.3
VAPO-Thinker-3B 2025.09		67.1
Qwen2.5-VL-7B 2026.04		67.1
Baseline 2026.03		66.3
Kimi-VL-16B 2025.10		66
OpenVLThinker-7B 2025.10		65.3
InternVL2.5-8B 2025.10		64.5
R1-OneVision-7B 2025.09		64.1
R1-Onevision-7B 2025.10		64.1
InternLM-XComposer2.5 2025.10		64
R1-VL-7B 2025.10		63.5
LLaVA-OneVision-7B 2025.10		62.6
Qwen2.5-VL-7B 2025.09		62.3
Qwen2.5-VL-3B 2026.04		62.3
Qwen2-VL-7B 2025.10		61.6
Qwen2-VL-7B 2026.04		61.2
CHARTOOL-3B 2026.04		60.5
Migician-7B 2026.04		60.2
HeadLens 2026.03		58.3
InternVL2-8B 2025.10		58.3
Baseline 2026.03		57.3
HeadLens 2026.03		26.3
Baseline 2026.03		23.9