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

Multimodal Reasoning on MMStar

82Accuracy

Masters

Updated 3d ago

Evaluation Results

Method
Masters 2025.12	82	-
Masters 2025.12	80.8	-
Masters 2025.12	79.7	-
Qwen3-VL-32B 2025.12	77.7	-
GPT-5 2025.12	75.7	-
GLM-4.5V 2025.12	75.3	-
InternVL3.5-38B 2025.12	75.3	-
Masters 2025.12	74.9	-
GPT-5-Mini 2025.12	74.1	-
ERNIE 5.0-Base 2026.02	74.07	-
Gemini-2.5-Pro 2025.12	73.6	-
InternVL3-78B 2025.12	72.5	-
Qwen2.5-VL-72B 2025.12	70.8	-
Qwen2.5-VL-72B 2025.12	70.8	-
GPT-4.1 2025.12	69.8	-
Claude-4-Sonnet 2025.12	69.4	-
Gemini-2.0-Flash 2025.12	69.4	-
Qwen2.5vl-Instruct 2025.12	66.5	-
LLaVA-OneVision 2025.12	66.1	-
LLaVA-OneVision-72B 2025.12	65.8	-
ReLaX-VL-7B 2025.12	65.5	-
Claude-3.5-Sonnet 2025.12	65.1	-
Claude-3.7-Sonnet 2025.12	65.1	-
Qwen2.5vl-Instruct 2025.12	64.8	-
GPT-4o 2025.12	64.7	-
Qwen2.5vl-Instruct 2025.12	64.3	-
MM-Eureka-7B 2025.12	64.3	-
VL-Rethinker-7B 2025.12	64.2	-
Active-Zero 2026.02	64	-
Qwen2.5-VL-7B 2025.12	63.9	-
NVLM-72B 2025.12	63.7	-
InternVL2.5-8B-VisualPRM 2025.12	63.4	-
Molmo-72B 2025.12	63.3	-
VisPlay 2026.02	62.93	-
VisionZero-Chart 2026.02	62.27	-
VisionZero-RealWorld 2026.02	62.27	-
VisionZero-CLEVR 2026.02	62.07	-
LLaVA-OneVision 2025.12	61.7	-
Base Model 2026.02	61.53	-
EvoLMM 2026.02	61.53	-
Intern2-VL-8B 2025.12	61.5	-
Vision-R1-7B 2025.12	60.9	-
Qwen2-VL-7B 2025.12	60.7	-
ReLaX-VL-3B 2025.12	60.7	-
R1-VL-7B 2025.12	60	-
R1-Onevision 2025.12	59.1	-
Gemini-1.5-Pro 2025.12	59.1	-
GPT-4V 2025.12	57.1	-
Qwen2.5-VL-3B-Instruct 2025.12	56.73	-
OpenVLThinker-7B 2025.12	56.3	-
Qwen2.5-VL-3B 2025.12	55.9	-
MobileNet-Qwen 2025.12	55.07	-
InternViT-Qwen 2025.12	54.93	-
Cambrian-1 2025.12	54.2	-
Active-Zero 2026.02	53.8	-
Base Model 2026.02	52.8	-
VisPlay 2026.02	51.6	-
InternViT-Liquid 2025.12	50.87	-
InternVL2-2B 2025.12	50.07	-
AutoNeural 2025.12	49.4	-
Qwen2-VL-2B 2025.12	47.8	-
InternVL2-1B 2025.12	45.47	-
URSA 2025.12	42.3	-
MGM-HD+CAL 2024.05	38.5	-
LLaVA-NeXT+CAL 2024.05	38.1	-
LLaVA-NeXT 2024.05	37.5	-
LLaVA-1.5+CAL 2024.05	35.7	-
MGM+CAL 2024.05	35.5	-
LLaVA-NeXT+CAL 2024.05	35.5	-
MGM-HD 2024.05	34.2	-
MGM+CAL 2024.05	33.7	-
LLaVA-NeXT 2024.05	33.7	-
LLaVA-1.5 2024.05	33.5	-
LLaVA-1.5+CAL 2024.05	32.9	-
MGM 2024.05	32.8	-
LLaVA-1.5 2024.05	32.4	-
MGM-HD 2024.05	31	-
MGM 2024.05	30.4	-
MGM-HD+CAL 2024.05	30.1	-
MGM 2024.05	25.5	-
MGM+CAL 2024.05	25.4	-
Qwen-VL-Max-0809 2024.10	-	69.2
GPT-4o-20240806 2024.10	-	64.7
InternVL2-Llama3-76B 2024.10	-	67.1
Claude3.5-Sonnet 2024.10	-	62.2
Gemini-1.5-Pro 2024.10	-	59.1
GPT-4v (1106, detail-high) 2024.10	-	50.4
Gemini-1.0-Pro 2024.10	-	38.6
Claude3-Sonnet 2024.10	-	44.2
Qwen-VL-Plus 2024.10	-	39.7
Qwen2-VL-2B 2024.10	-	47.5
H2OVL-Mississippi-2B 2024.10	-	49.6
InternVL2-2B 2024.10	-	49.8
Phi-3-Vision 2024.10	-	47.7
MiniMonkey 2024.10	-	48.1
MiniCPM-V-2 2024.10	-	39.1
InternVL2-1B 2024.10	-	45.6
PaliGemma-3B-mix-448 2024.10	-	48.3
H2OVL-Mississippi-0.8B 2024.10	-	39.1
DeepSeek-VL-1.3B 2024.10	-	39.9

Showing 100 of 130 rows