Kimi K3.1 Pro
Kimi K3.1 Pro scores 70.6 out of 100 overall, which puts it in a tie for rank 6–9 of 36 models on the AI Vision Index. It is strongest at Diagrams (91) and weakest at Counting (53). It beats the average of all 36 models in all six categories. It was tested on a smaller sample than most models, so its margin of error is wider and its single-test scores are rough.
Overall score70.695% interval 66.6–75.0
Rank6–9of 36 models (tied)
Cost per 1,000 answers$71.79
Seconds per answer23.2 s
Thinking effortdefault
Maker Moonshot
WeightsOpen weights
AccessAPI
Tested
Reading72
Counting53
Space, 3D and motion62
Spotting changes and errors77
Diagrams91
Not being fooled68
The bar is the score; the light band is the 95% interval. The overall score is the average of the six.
Reading
Reading a clock80
Reading a meter87
Eye chart60
Reading a label87
Reading a chart67
Test strips47
Adding up coins60
Traffic light from a car90
Counting
Counting objects47
Counting stacked goods20
Counting on a conveyor93
Space, 3D and motion
Which is nearer100
Which way it faces69
Same or mirror image0
Will it topple40
Match the render56
Which bay can I park in?100
Follow the hidden object72
Spotting changes and errors
Spot the change87
Shelf against plan87
Rendering glitch36
Check the invoice100
Diagrams
Follow the cable87
Follow the pipe87
Missing dimension100
Not being fooled
Absent objects100
Saying "I can't tell"100
Optical illusions50
Real or AI-made20
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 Sonnet 5.5 71.0 Compare
- Qwen3.8 Max 70.6 Compare
Badge for your site
HTML
<a href="https://aivisionindex.com/models/kimi-k3-1-pro/"><img src="https://aivisionindex.com/badge/kimi-k3-1-pro.svg" alt="Kimi K3.1 Pro: ranked #6–9 of 36 on the AI Vision Index"></a>Markdown
[](https://aivisionindex.com/models/kimi-k3-1-pro/)Results are CC BY 4.0, so you are free to reuse them if you credit the AI Vision Index. All results (JSON)