adjustable glasses illustration

eye chart illustration

clipboard illustration

lenses illustration

GPT-4.1 vs.ChatGPT-4o

Passed:

GPT-4.1 73.2% (60/82)

ChatGPT-4o 78.0% (64/82)

Average request time:

GPT-4.1 14.91s

ChatGPT-4o 13.60s

Summary

Annotation Understanding

GPT-4.1 1 / 2

ChatGPT-4o 1 / 2

CAPTCHA

GPT-4.1 0 / 2

ChatGPT-4o 2 / 2

Color Identification

GPT-4.1 1 / 1

ChatGPT-4o 1 / 1

Counting nodes

GPT-4.1 0 / 1

ChatGPT-4o 0 / 1

Defect Detection

GPT-4.1 13 / 15

ChatGPT-4o 14 / 15

Document Understanding

GPT-4.1 5 / 9

ChatGPT-4o 5 / 9

Find an author

GPT-4.1 0 / 1

ChatGPT-4o 0 / 1

Localization

GPT-4.1 1 / 1

ChatGPT-4o 1 / 1

OCR

GPT-4.1 9 / 9

ChatGPT-4o 9 / 9

Object Counting

GPT-4.1 2 / 10

ChatGPT-4o 2 / 10

Object Detection

GPT-4.1 2 / 2

ChatGPT-4o 2 / 2

Object Measurement

GPT-4.1 1 / 2

ChatGPT-4o 1 / 2

Object Understanding

GPT-4.1 9 / 10

ChatGPT-4o 9 / 10

Receipt Reading

GPT-4.1 1 / 1

ChatGPT-4o 1 / 1

Sign Understanding

GPT-4.1 2 / 2

ChatGPT-4o 2 / 2

Spatial Relations

GPT-4.1 14 / 17

ChatGPT-4o 16 / 17

Web Action Understanding

GPT-4.1 4 / 4

ChatGPT-4o 4 / 4

Contribute a Prompt

Lightbulb

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