LFM2.5-VL 1.6B
LFM2.5-VL 1.6B scores 12.4 out of 100 overall, which puts it in a tie for rank 34–36 of 36 models on the AI Vision Index. It is strongest at Not being fooled (29) and weakest at Space, 3D and motion (3). It beats the average of all 36 models in none of the six categories.
Overall score12.495% interval 11.4–14.2
Rank34–36of 36 models (tied)
Cost per 1,000 answers$0 (local)
Seconds per answer0.7 s on a laptop GPU
Thinking effortdefault
Maker Liquid
WeightsOpen weights
AccessRuns locally
Tested
Reading18
Counting6
Space, 3D and motion3
Spotting changes and errors15
Diagrams4
Not being fooled29
The bar is the score; the light band is the 95% interval. The overall score is the average of the six.
Reading
Reading a clock2
Reading a meter6
Eye chart0
Reading a label41
Reading a chart37
Test strips24
Adding up coins0
Traffic light from a car34
Counting
Counting objects11
Counting stacked goods2
Counting on a conveyor4
Space, 3D and motion
Which is nearer12
Which way it faces0
Same or mirror image5
Will it topple1
Match the render1
Which bay can I park in?0
Follow the hidden object0
Spotting changes and errors
Spot the change16
Shelf against plan8
Rendering glitch30
Check the invoice5
Diagrams
Follow the cable7
Follow the pipe3
Missing dimension2
Not being fooled
Absent objects55
Saying "I can't tell"55
Optical illusions0
Real or AI-made5
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
Badge for your site
HTML
<a href="https://aivisionindex.com/models/lfm2-5-vl-1-6b/"><img src="https://aivisionindex.com/badge/lfm2-5-vl-1-6b.svg" alt="LFM2.5-VL 1.6B: ranked #34–36 of 36 on the AI Vision Index"></a>Markdown
[](https://aivisionindex.com/models/lfm2-5-vl-1-6b/)Results are CC BY 4.0, so you are free to reuse them if you credit the AI Vision Index. All results (JSON)