Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Chart parsing on ChartQA
Loading...
86.8
Accuracy
MAESTRO
68.08
72.94
77.8
82.66
May 21, 2026
Accuracy
Updated 12d ago
Evaluation Results
Method
Method
Links
Accuracy
MAESTRO
Model Type=Ours
2026.05
86.8
VTOOL-R1
Model Type=Think with...
2026.05
86.7
Thyme
Model Type=Think with...
2026.05
86.1
GLM-4.6V
Model Type=Open-Source...
2026.05
85
Gemini-2.5-Pro
Model Type=Closed-Source
2026.05
83.6
GPT-4o
Model Type=Closed-Source
2026.05
81.4
VTS-V
Model Type=Think with...
2026.05
81.2
Gemini-2.5-Flash
Model Type=Closed-Source
2026.05
79.6
Kimi-K2.5
Model Type=Open-Source...
2026.05
79.4
VisionReasoner
Model Type=Think with...
2026.05
79.2
Visual-ARFT
Model Type=Think with...
2026.05
79
MathCoder-VL
Model Type=Think with...
2026.05
78.8
Qwen3-VL-32B
Model Type=Open-Source...
2026.05
77.8
Direct Answering
Model Type=Open-Source...
2026.05
76.8
GPT-5
Model Type=Closed-Source
2026.05
76.7
PixelReasoner
Model Type=Think with...
2026.05
76.2
Untrained Model
Model Type=Open-Source...
2026.05
74.8
DeepEyes-v2
Model Type=Think with...
2026.05
72.2
DeepEyes
Model Type=Think with...
2026.05
69.4
Chain-of-Focus
Model Type=Think with...
2026.05
68.8
Feedback
Search any
task
Search any
task