GPT-6 Luna
GPT-6 Luna scores 60.7 out of 100 overall, which puts it in a tie for rank 10–18 of 36 models on the AI Vision Index. It is strongest at Spotting changes and errors (73) and weakest at Counting (35). It beats the average of all 36 models in all six categories.
Overall score60.795% interval 59.3–62.2
Rank10–18of 36 models (tied)
Cost per 1,000 answers$0.74
Seconds per answer9.2 s
Thinking effortmedium
Maker OpenAI
WeightsClosed weights
AccessAPI
Tested
Reading59
Counting35
Space, 3D and motion54
Spotting changes and errors73
Diagrams71
Not being fooled73
The bar is the score; the light band is the 95% interval. The overall score is the average of the six.
Reading
Reading a clock75
Reading a meter70
Eye chart40
Reading a label66
Reading a chart46
Test strips62
Adding up coins44
Traffic light from a car70
Counting
Counting objects38
Counting stacked goods33
Counting on a conveyor33
Space, 3D and motion
Which is nearer65
Which way it faces49
Same or mirror image8
Will it topple51
Match the render65
Which bay can I park in?92
Follow the hidden object47
Spotting changes and errors
Spot the change77
Shelf against plan78
Rendering glitch44
Check the invoice91
Diagrams
Follow the cable69
Follow the pipe58
Missing dimension86
Not being fooled
Absent objects90
Saying "I can't tell"99
Optical illusions77
Real or AI-made24
Scores are corrected for guessing: 0 is no better than a guess, 100 is always right. All models on every test
Neighbours on the leaderboard
- Gemini 3.8 Flash 61.9 Compare
- Qwen3.8-27B 59.0 Compare
Badge for your site
HTML
<a href="https://aivisionindex.com/models/gpt-6-luna/"><img src="https://aivisionindex.com/badge/gpt-6-luna.svg" alt="GPT-6 Luna: ranked #10–18 of 36 on the AI Vision Index"></a>Markdown
[](https://aivisionindex.com/models/gpt-6-luna/)Results are CC BY 4.0, so you are free to reuse them if you credit the AI Vision Index. All results (JSON)