List your company

Multimodal Mathematical Reasoning on MathVerse-V

81.2AccuracyGemini-2.5-Pro-Thinking
28.274442.014755.75569.4953Dec 19, 2025Jan 5, 2026Jan 22, 2026Feb 9, 2026Feb 26, 2026Mar 15, 2026Apr 2, 2026

Evaluation results

MethodLinks
Gemini-2.5-Pro-ThinkingAccess Type=Closed-Source2025.1281.2
Seed-1.5-thinkingAccess Type=Closed-Source2025.1280.4
Qwen2.5-VL-7B + PGPOModel Size=7B, Optimization Strategy=PGPO2026.0466.41
Claude-4-SonnetAccess Type=Closed-Source2025.1266.1
Qwen2.5-VL-7B + VPPOModel Size=7B, Optimization Strategy=VPPO2026.0465.54
VL-Rethinker-7BModel Size=7B2026.0464.96
Qwen2.5-VL-7B + PAPOModel Size=7B, Optimization Strategy=PAPO2026.0464.92
Qwen2.5-VL-7B + GRPOModel Size=7B, Optimization Strategy=GRPO2026.0464.78
Qwen2.5-VL-7B + DAPOModel Size=7B, Optimization Strategy=DAPO2026.0464.38
R1-ShareVL-7BModel Size=7B2026.0464.34
NoisyRollout-7BModel Size=7B2026.0463.78
MM-Eureka-7BModel Size=7B2026.0462.41
GPT-4.1Access Type=Closed-Source2025.1259.8
Qwen2.5-VL-3B + PGPOModel Size=3B, Optimization Strategy=PGPO2026.0457.24
Qwen2.5-VL-3B + VPPOModel Size=3B, Optimization Strategy=VPPO2026.0457.02
Qwen2.5-VL-3B + PAPOModel Size=3B, Optimization Strategy=PAPO2026.0456.9
Qwen2.5-VL-3B + DAPOModel Size=3B, Optimization Strategy=DAPO2026.0454.89
Qwen2.5-VL-3B + GRPOModel Size=3B, Optimization Strategy=GRPO2026.0454.36
Revisual-R1Access Type=Open-Source, Model Scale=7B2025.1253.6
VAPO-ThinkerAccess Type=Open-Source, Model Scale=7B2025.1253.3
OursAccess Type=Open-Source, Model Scale=7B2025.1253.3
Look-BackAccess Type=Open-Source, Model Scale=7B2025.1251.8
Vision-G1Access Type=Open-Source, Model Scale=7B2025.1250
MM-EurekaAccess Type=Open-Source, Model Scale=7B2025.1249.6
VL-RethinkerAccess Type=Open-Source, Model Scale=7B2025.1249.1
VLAA-ThinkingAccess Type=Open-Source, Model Scale=7B2025.1248.2
ThinkLite-VLAccess Type=Open-Source, Model Scale=7B2025.1245.3
MathBookAccess Type=Open-Source, Model Scale=7B2025.1245.2
MMR1-Math-v0Access Type=Open-Source, Model Scale=7B2025.1245.1
Qwen2.5-VLAccess Type=Open-Source, Model Scale=7B2025.1242.9
OpenVLThinkerAccess Type=Open-Source, Model Scale=7B2025.1240.7
Qwen2.5-VL-7BModel Size=7B2026.0433.63
Qwen2.5-VL-3BModel Size=3B2026.0430.31