GPT-6 Astra
GPT-6 Astra scores 83.7 out of 100 overall, which puts it at rank 1 of 36 models on the AI Vision Index. It is strongest at Diagrams (98) and weakest at Not being fooled (71). It beats the average of all 36 models in all six categories.
Overall score83.795% interval 81.7–85.7
Rank1of 36 models
Cost per 1,000 answers$38.32
Seconds per answer5.5 s
Thinking effortmedium
Maker OpenAI
WeightsClosed weights
AccessAPI
Tested
Reading88
Counting78
Space, 3D and motion84
Spotting changes and errors84
Diagrams98
Not being fooled71
The bar is the score; the light band is the 95% interval. The overall score is the average of the six.
Reading
Reading a clock98
Reading a meter98
Eye chart98
Reading a label78
Reading a chart91
Test strips75
Adding up coins80
Traffic light from a car89
Counting
Counting objects52
Counting stacked goods80
Counting on a conveyor100
Space, 3D and motion
Which is nearer60
Which way it faces85
Same or mirror image75
Will it topple85
Match the render87
Which bay can I park in?97
Follow the hidden object97
Spotting changes and errors
Spot the change92
Shelf against plan88
Rendering glitch55
Check the invoice100
Diagrams
Follow the cable92
Follow the pipe100
Missing dimension100
Not being fooled
Absent objects94
Saying "I can't tell"96
Optical illusions81
Real or AI-made15
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
- GPT-6.1 Sol 81.3 Compare
Badge for your site
HTML
<a href="https://aivisionindex.com/models/gpt-6-astra/"><img src="https://aivisionindex.com/badge/gpt-6-astra.svg" alt="GPT-6 Astra: ranked #1 of 36 on the AI Vision Index"></a>Markdown
[](https://aivisionindex.com/models/gpt-6-astra/)Results are CC BY 4.0, so you are free to reuse them if you credit the AI Vision Index. All results (JSON)