Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Mathematical Reasoning on MathVista (Pass@1 accuracy)
Loading...
73.8
Pass@1 Accuracy
Qwen2.5-VL-7B-Instruct + RFT
52.376
57.938
63.5
69.062
Apr 11, 2026
Pass@1 Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Pass@1 Accuracy
Qwen2.5-VL-7B-Instruct + RFT
Model Source Category=...
2026.04
73.8
InternVL3-9B
Model Source Category=...
2026.04
71.5
Qwen2.5-VL-7B-Instruct + cold start
Model Source Category=...
2026.04
71.1
Gemini-1.5-Pro
Model Source Category=...
2026.04
68.7
InternVL2.5-26B
Model Source Category=...
2026.04
68.2
Qwen2.5-VL-7B-Instruct
Model Source Category=...
2026.04
68.2
Qwen2.5-VL-7B-Instruct
Model Source Category=...
2026.04
68.1
Qwen2.5-VL-7B-Instruct
Model Source Category=...
2026.04
68.1
LLaVA-OneVision-72B
Model Source Category=...
2026.04
67.1
Claude-3.5-Sonnet
Model Source Category=...
2026.04
64.7
InternVL2.5-8B
Model Source Category=...
2026.04
64.5
Qwen2-VL-7B
Model Source Category=...
2026.04
62.3
MiniCPM-V2.6
Model Source Category=...
2026.04
60.8
GPT-4o-20240513
Model Source Category=...
2026.04
60
Cambrian-34B
Model Source Category=...
2026.04
53.2
Feedback
Search any
task
Search any
task