adjustable glasses illustration

eye chart illustration

clipboard illustration

lenses illustration

Gemma 4 31B vs.Phi 4 Multimodal

Passed:

Gemma 4 31B 72.2% (39/54)

Phi 4 Multimodal 49.4% (41/83)

Average request time:

Gemma 4 31B 15.30s

Phi 4 Multimodal 15.92s

Summary

Annotation Understanding

Gemma 4 31B 0 / 1

Phi 4 Multimodal 1 / 2

CAPTCHA

Gemma 4 31B 2 / 2

Phi 4 Multimodal 0 / 2

Color Identification

Gemma 4 31B 0 / 0

Phi 4 Multimodal 0 / 1

Defect Detection

Gemma 4 31B 6 / 8

Phi 4 Multimodal 11 / 15

Document Understanding

Gemma 4 31B 7 / 7

Phi 4 Multimodal 7 / 10

Localization

Gemma 4 31B 0 / 1

Phi 4 Multimodal 0 / 1

OCR

Gemma 4 31B 4 / 4

Phi 4 Multimodal 5 / 9

Object Counting

Gemma 4 31B 2 / 8

Phi 4 Multimodal 0 / 11

Object Detection

Gemma 4 31B 1 / 2

Phi 4 Multimodal 2 / 2

Object Measurement

Gemma 4 31B 0 / 0

Phi 4 Multimodal 0 / 1

Object Understanding

Gemma 4 31B 4 / 7

Phi 4 Multimodal 5 / 11

Receipt Reading

Gemma 4 31B 1 / 1

Phi 4 Multimodal 1 / 1

Sign Understanding

Gemma 4 31B 1 / 1

Phi 4 Multimodal 2 / 2

Spatial Relations

Gemma 4 31B 10 / 13

Phi 4 Multimodal 7 / 17

Web Action Understanding

Gemma 4 31B 4 / 4

Phi 4 Multimodal 2 / 4

Contribute a Prompt

Lightbulb

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