Gemini 4 Argon
Gemini 4 Argon scores 77.2 out of 100 overall, which puts it in a tie for rank 4–5 of 36 models on the AI Vision Index. It is strongest at Diagrams (90) and weakest at Counting (58). It beats the average of all 36 models in all six categories.
Overall score77.295% interval 75.9–78.5
Rank4–5of 36 models (tied)
Cost per 1,000 answers$6.52
Seconds per answer7.5 s
Thinking effortdefault
Maker Google
WeightsClosed weights
AccessAPI
Tested
Reading80
Counting58
Space, 3D and motion72
Spotting changes and errors89
Diagrams90
Not being fooled74
The bar is the score; the light band is the 95% interval. The overall score is the average of the six.
Reading
Reading a clock96
Reading a meter90
Eye chart82
Reading a label89
Reading a chart58
Test strips72
Adding up coins66
Traffic light from a car84
Counting
Counting objects37
Counting stacked goods54
Counting on a conveyor83
Space, 3D and motion
Which is nearer90
Which way it faces97
Same or mirror image24
Will it topple61
Match the render82
Which bay can I park in?98
Follow the hidden object52
Spotting changes and errors
Spot the change94
Shelf against plan94
Rendering glitch73
Check the invoice94
Diagrams
Follow the cable85
Follow the pipe95
Missing dimension91
Not being fooled
Absent objects96
Saying "I can't tell"100
Optical illusions40
Real or AI-made61
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
- Claude Fable 5.2 80.7 Compare
- Claude Opus 5.5 76.5 Compare
Badge for your site
HTML
<a href="https://aivisionindex.com/models/gemini-4-argon/"><img src="https://aivisionindex.com/badge/gemini-4-argon.svg" alt="Gemini 4 Argon: ranked #4–5 of 36 on the AI Vision Index"></a>Markdown
[](https://aivisionindex.com/models/gemini-4-argon/)Results are CC BY 4.0, so you are free to reuse them if you credit the AI Vision Index. All results (JSON)