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

Mathematical Reasoning on MathVision (Accuracy)

83.9Accuracy

Qwen3.5

Updated 18d ago

Evaluation Results

Method	Links
Qwen3.5 2026.05		83.9
Gemini3-pro 2026.05		83.4
SenseNova-U1 2026.05		79.63
Qwen3.5 2026.05		78.9
SenseNova-U1 2026.05		75.82
Qwen3-VL-235B-Thinking 2026.05		74.6
GPT-5-20250807 2026.05		72
Gemma4 2026.05		68.95
Doubao-Seed-1.6 2026.05		67.8
Qwen3VL 2026.05		65.7
GLM-4.5V 2026.05		65.6
LongCat-Next 2026.05		64.7
Qwen3VL 2026.05		62.7
OST 2026.05		58.5
DEITA 2026.05		58.3
Qwen3-VL-8B-Instruct + LLM-as-a-Judge 2026.05		58.2
Qwen3-VL-8B-Instruct 2026.05		57.8
Qwen3-VL-8B-Instruct + Random 2026.05		57.1
Kimi-vl-A3B-thinking 2026.05		56.8
OST 2026.05		54.8
Qwen3-VL-8B-Instruct + Full SFT 2026.05		54.8
Qwen3-VL-4B-Instruct + LLM-as-a-Judge 2026.05		54.6
DEITA 2026.05		54.5
Qwen3-VL-4B-Instruct 2026.05		54.1
Qwen3-VL-4B-Instruct + Random 2026.05		54
Qwen3-VL-4B-Instruct + Full SFT 2026.05		53.8
Qwen3-VL-8B 2026.07		51.61
Gemini-2.0-Flash 2025.11		47.8
Qwen3-VL-4B-Instruct / GRPO 2026.07		46.7
Qwen3-VL-4B-Instruct / H-OPD (8B) 2026.07		46.7
Qwen3-VL-4B-Instruct / ExOPD (8B) 2026.07		46.4
Qwen3-VL-4B-Instruct / OPD (8B) 2026.07		46.1
Qwen3-VL-8B 2026.07		45.4
Qwen3-VL-4B 2026.07		44.7
Qwen3-VL-4B-Instruct / SFT (8B) 2026.07		42.1
Claude-3.5 Sonnet 2025.11		41.3
Qwen3-VL-2B-Instruct / H-OPD (8B) 2026.07		36.8
Qwen3-VL-8B 2026.07		36.78
MM-Eureka-Qwen-32B 2025.11		35.5
MM-Eureka-Qwen-32B 2025.11		35.2
Qwen3-VL-2B-Instruct + LLM-as-a-Judge 2026.05		35
QvQ-72B-Preview 2025.11		34.9
DEITA 2026.05		34.9
OST 2026.05		34.8
TemplateRL 2025.05		34.6
Qwen3-VL-2B-Instruct 2026.05		34.5
Qwen3-VL-2B-Instruct + Random 2026.05		34.3
Qwen3-VL-2B-Instruct + Full SFT 2026.05		34.1
MMR1-Math-v0-7B 2025.11		33.9
Qwen3-VL-2B-Instruct / H-OPD (4B) 2026.07		33.5
VL-Rethinker-7B 2025.11		33.2
MMR1-Math-v0-7B 2025.11		32.6
Qwen3-VL-8B 2026.07		31.78
Qwen3-VL-2B-Instruct / ExOPD (4B) 2026.07		31.6
MM-Eureka-Qwen-7B 2025.11		31.5
GPT-4o 2025.11		31.1
GRPO 2025.05		30.9
Qwen3-VL-2B-Instruct / OPD (4B) 2026.07		30.9
MM-Eureka-Qwen-7B 2025.11		30.6
Qwen3-VL-2B-Instruct / ExOPD (8B) 2026.07		30.6
VL-Rethinker-7B 2025.11		30.3
Qwen3-VL-2B-Instruct / GRPO 2026.07		30.3
Qwen3-VL-2B-Instruct / OPD (8B) 2026.07		30.3
InternVL3-8B 2025.11		28.6
Qwen3-VL-2B 2026.07		28.6
Qwen3-VL-2B-Instruct / SFT (8B) 2026.07		28.2
CoT 2025.05		27.7
GPT-4o-mini 2025.11		27.3
Qwen3-VL-2B-Instruct / SFT (4B) 2026.07		26.8
Qwen3-VL-8B 2026.07		26.35
+BETAPRM 2026.05		25.66
+BETAPRM 2026.05		25.66
Qwen2.5-VL-7B 2025.11		25.5
+BETAPRM 2026.05		24.34
+BETAPRM 2026.05		24.34
Qwen3-VL-8B 2026.07		23.98
+Standard PRM 2026.05		23.03
+Standard PRM 2026.05		22.69
Kimi-VL-16B 2025.11		21.8
+Standard PRM 2026.05		21.38
+Standard PRM 2026.05		21.38
+Base (w/o training) 2026.05		20.72
+Base (w/o training) 2026.05		19.74
Qwen3-VL-8B 2026.07		18.82
+Base (w/o training) 2026.05		18.75
Single Pass 2026.05		18.08
LLaVA-OV-7B 2025.11		17
Qwen2.5-VL-3B 2025.11		16.1
+Base (w/o training) 2026.05		15.46