Our new X account is live! Follow @wizwand_team for updates

Multimodal Mathematical Reasoning on MathVista MINI

0.821Accuracy

Qwen3-VL-8B-Instruct + CARE

Updated 3d ago

Evaluation Results

Method	Links
Qwen3-VL-8B-Instruct + CARE 2025.12		0.821
MiMo-VL-7B-SFT 2025.12		0.818
MiMo-VL-7B-RL 2025.12		0.815
Qwen3-VL-8B-Thinking 2025.12		0.814
Keye-VL-1.5-8B 2025.12		0.807
Qwen3-VL-4B-Thinking 2025.12		0.795
Qwen3-VL-8B-Instruct 2025.12		0.772
M2-Reasoning 2025.12		0.75
Qwen2.5-VL-7B + CARE 2025.12		0.747
InternVL3.5-8B-Instruct 2025.12		0.742
InternVL3.5-8B-MPO 2025.12		0.742
Qwen2.5-VL-7B + GSPO 2025.12		0.741
VideoAuto-R1 2026.01		0.737
Qwen3-VL-4B-Instruct 2025.12		0.737
Vision-R1-7B 2025.12		0.735
GPT-5-Nano 2025.12		0.731
Qwen2.5-VL-7B + DAPO 2025.12		0.726
Gemini-2.0-Pro 2025.12		0.713
DeepEyes 2025.12		0.701
LLaVA-OneVision-1.5 8B 2025.12		0.696
Qwen2.5-VL-7B 2026.01		0.694
Qwen2.5-VL-7B + GRPO 2025.12		0.689
Qwen2.5-VL-7B 2025.12		0.686
Claude-Sonnet-3.7 2025.12		0.668
Qwen2.5-VL-3B + CARE 2025.12		0.665
GPT-4o 2025.12		0.638
Qwen2.5-VL-3B-Instruct 2025.12		0.62
Qwen2.5-VL-3B 2025.12		0.62
MobileNet-Qwen 2025.12		0.606
InternViT-Qwen 2025.12		0.575
AutoNeural 2025.12		0.531
InternViT-Liquid 2025.12		0.526
InternVL2-2B 2025.12		0.491
Qwen2-VL-2B 2025.12		0.471
InternVL2-1B 2025.12		0.403