List your company
ResearchFormerly WizWand

Mathematical multi-modal reasoning on WeMath

85.11Pass@1Qwen3-VL-8B-DeepVision
21.87838.29454.7171.126Jun 5, 2025Jul 31, 2025Sep 26, 2025Nov 22, 2025Jan 18, 2026Mar 16, 2026May 12, 2026
Updated 6mo ago

Evaluation results

MethodLinks
Qwen3-VL-8B-DeepVisionSeries=Qwen3-VL-8B, Training Algorithm=GSPO2026.0285.11
Qwen3-VL-8B-ThinkingSeries=Qwen3-VL-8B2026.0284.54
Gemini-2.5-Flash-LiteModel Category=Closed-source Models2026.0283.85
MiMo-VL-7B-OpenMMReasonerSeries=MiMo-VL-7B2026.0283.45
MiMo-VL-7B-DeepVisionSeries=MiMo-VL-7B, Training Algorithm=GSPO2026.0282.98
Qwen3-VL-8B-InstructSeries=Qwen3-VL-8B2026.0279.36
MiMo-VL-7B-MM-EurekaSeries=MiMo-VL-7B2026.0279.08
GPT-5-Nano-HighModel Category=Closed-source Models2026.0278.62
MiMo-VL-7B-MathBookSeries=MiMo-VL-7B2026.0277.18
MiMo-VL-7B-RL-2508Series=MiMo-VL-7B2026.0276.95
MiMo-VL-7B-SFT-2508Series=MiMo-VL-7B2026.0274.42
Self-distill Masking-KD-8BModel Size=~8B, Decoding Strategy=greedy, Max Length=4096, Distillation=Self-distill2026.0571.72
Masking-KD-4BModel Size=~4B, Decoding Strategy=greedy, Max Length=4096, Distillation=Distilled from 8B2026.0571.03
Qwen3-VL-8B-ThinkingModel Size=~8B, Decoding Strategy=greedy, Max Length=40962026.0566.15
Ovis2-8BModel Size=~8B, Decoding Strategy=greedy, Max Length=40962026.0564.66
Masking-KD-2BModel Size=~2B, Decoding Strategy=greedy, Max Length=4096, Distillation=Distilled from 8B2026.0563.79
Ovis2-4BModel Size=~4B, Decoding Strategy=greedy, Max Length=40962026.0560.29
InternVL3.5-8BModel Size=~8B, Decoding Strategy=greedy, Max Length=40962026.0556.61
Ovis2-2BModel Size=~2B, Decoding Strategy=greedy, Max Length=40962026.0551.95
InternVL3-8BModel Size=~8B, Decoding Strategy=greedy, Max Length=40962026.0551.32
Qwen2.5-VL-3BModel Size=~4B, Decoding Strategy=greedy, Max Length=40962026.0549.66
Qwen3-VL-4B-ThinkingModel Size=~4B, Decoding Strategy=greedy, Max Length=40962026.0549.37
Qwen2.5-VL-7BModel Size=~8B, Decoding Strategy=greedy, Max Length=40962026.0548.74
InternVL3-2BModel Size=~2B, Decoding Strategy=greedy, Max Length=40962026.0547.93
InternVL3.5-4BModel Size=~4B, Decoding Strategy=greedy, Max Length=40962026.0545.46
MiMo-VL-8BModel Size=~8B, Decoding Strategy=greedy, Max Length=40962026.0542.41
Qwen2.5-VL-7b-RAPBackbone=Qwen2.5-VL-7b, Sample=5,159, Time (h)=52.82025.0642
Qwen2.5-VL-3b-RAPBackbone=Qwen2.5-VL-3b, Sample=4,374, Time (h)=322025.0629.33
Qwen3-VL-2B-ThinkingModel Size=~2B, Decoding Strategy=greedy, Max Length=40962026.0525.17
InternVL3.5-2BModel Size=~2B, Decoding Strategy=greedy, Max Length=40962026.0524.31