adjustable glasses illustration

eye chart illustration

clipboard illustration

lenses illustration

Qwen 2.5 VL 7B vs.Gemini 1.5 Pro

Passed:

Qwen 2.5 VL 7B 55.6% (45/81)

Gemini 1.5 Pro 63.9% (53/83)

Average request time:

Qwen 2.5 VL 7B 45.63s

Gemini 1.5 Pro 14.78s

Summary

Annotation Understanding

Qwen 2.5 VL 7B 1 / 2

Gemini 1.5 Pro 1 / 2

CAPTCHA

Qwen 2.5 VL 7B 2 / 2

Gemini 1.5 Pro 2 / 2

Color Identification

Qwen 2.5 VL 7B 0 / 1

Gemini 1.5 Pro 0 / 1

Counting nodes

Qwen 2.5 VL 7B 0 / 0

Gemini 1.5 Pro 0 / 1

Defect Detection

Qwen 2.5 VL 7B 9 / 15

Gemini 1.5 Pro 11 / 15

Document Understanding

Qwen 2.5 VL 7B 8 / 10

Gemini 1.5 Pro 8 / 10

Localization

Qwen 2.5 VL 7B 1 / 1

Gemini 1.5 Pro 1 / 1

OCR

Qwen 2.5 VL 7B 6 / 9

Gemini 1.5 Pro 8 / 9

Object Counting

Qwen 2.5 VL 7B 1 / 9

Gemini 1.5 Pro 2 / 10

Object Detection

Qwen 2.5 VL 7B 1 / 2

Gemini 1.5 Pro 1 / 2

Object Measurement

Qwen 2.5 VL 7B 0 / 2

Gemini 1.5 Pro 0 / 2

Object Understanding

Qwen 2.5 VL 7B 5 / 11

Gemini 1.5 Pro 6 / 11

Receipt Reading

Qwen 2.5 VL 7B 0 / 1

Gemini 1.5 Pro 1 / 1

Sign Understanding

Qwen 2.5 VL 7B 2 / 2

Gemini 1.5 Pro 2 / 2

Spatial Relations

Qwen 2.5 VL 7B 11 / 17

Gemini 1.5 Pro 12 / 17

Web Action Understanding

Qwen 2.5 VL 7B 4 / 4

Gemini 1.5 Pro 4 / 4

Contribute a Prompt

Lightbulb

Have an idea for a prompt? Open a pull request on the project repository!