GPT-6.1 Sol
GPT-6.1 Sol scores 81.3 out of 100 overall, which puts it in a tie for rank 2–3 of 36 models on the AI Vision Index. It is strongest at Diagrams (96) and weakest at Not being fooled (72). It beats the average of all 36 models in all six categories.
Overall score81.395% interval 80.0–82.5
Rank2–3of 36 models (tied)
Cost per 1,000 answers$7.94
Seconds per answer6.1 s
Thinking effortmedium
Maker OpenAI
WeightsClosed weights
AccessAPI
Tested
Reading82
Counting76
Space, 3D and motion80
Spotting changes and errors81
Diagrams96
Not being fooled72
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 meter91
Eye chart88
Reading a label78
Reading a chart80
Test strips70
Adding up coins74
Traffic light from a car80
Counting
Counting objects50
Counting stacked goods81
Counting on a conveyor98
Space, 3D and motion
Which is nearer80
Which way it faces73
Same or mirror image61
Will it topple74
Match the render84
Which bay can I park in?97
Follow the hidden object92
Spotting changes and errors
Spot the change87
Shelf against plan85
Rendering glitch53
Check the invoice99
Diagrams
Follow the cable91
Follow the pipe98
Missing dimension98
Not being fooled
Absent objects89
Saying "I can't tell"100
Optical illusions80
Real or AI-made19
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 Astra 83.7 Compare
- Claude Fable 5.2 80.7 Compare
Badge for your site
HTML
<a href="https://aivisionindex.com/models/gpt-6-1-sol/"><img src="https://aivisionindex.com/badge/gpt-6-1-sol.svg" alt="GPT-6.1 Sol: ranked #2–3 of 36 on the AI Vision Index"></a>Markdown
[](https://aivisionindex.com/models/gpt-6-1-sol/)Results are CC BY 4.0, so you are free to reuse them if you credit the AI Vision Index. All results (JSON)