Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Visual Understanding on MME RealWorld
Loading...
72.7
Pass@1 Exact Match
SenseNova-MARS-32B
56.164
60.457
64.75
69.043
Dec 30, 2025
Pass@1 Exact Match
Updated 3d ago
Evaluation Results
Method
Method
Links
Pass@1 Exact Match
SenseNova-MARS-32B
Method Category=Agenti...
2025.12
72.7
Skywork-R1V4
Method Category=Agenti...
2025.12
71.4
SenseNova-MARS-8B
Method Category=Agenti...
2025.12
67.9
Mini o3
Method Category=Agenti...
2025.12
65.5
DeepEyesV2
Method Category=Agenti...
2025.12
64.9
Thyme
Method Category=Agenti...
2025.12
64.8
Pixel-Reasoner
Method Category=Agenti...
2025.12
64.4
DeepEyes
Method Category=Agenti...
2025.12
64.1
GPT-4o
Method Category=Direct...
2025.12
62.8
Qwen3-VL-8B-Instruct
Method Category=Direct...
2025.12
61.9
Qwen2.5-VL-32B-Instruct
Method Category=Direct...
2025.12
59.1
LLaVA-onevison
Method Category=Direct...
2025.12
57.4
Qwen2.5-VL-7B-Instruct
Method Category=Direct...
2025.12
56.8
Feedback
Search any
task
Search any
task