adjustable glasses illustration

eye chart illustration

clipboard illustration

lenses illustration

Cohere Aya Vision 32B vs.Gemma 3 4B

Passed:

Cohere Aya Vision 32B 56.6% (47/83)

Gemma 3 4B 42.2% (35/83)

Average request time:

Cohere Aya Vision 32B 18.49s

Gemma 3 4B 15.86s

Summary

Annotation Understanding

Cohere Aya Vision 32B 0 / 2

Gemma 3 4B 0 / 2

CAPTCHA

Cohere Aya Vision 32B 0 / 2

Gemma 3 4B 2 / 2

Color Identification

Cohere Aya Vision 32B 0 / 1

Gemma 3 4B 0 / 1

Counting nodes

Cohere Aya Vision 32B 0 / 1

Gemma 3 4B 0 / 1

Defect Detection

Cohere Aya Vision 32B 10 / 15

Gemma 3 4B 9 / 15

Document Understanding

Cohere Aya Vision 32B 6 / 10

Gemma 3 4B 6 / 10

Localization

Cohere Aya Vision 32B 1 / 1

Gemma 3 4B 0 / 1

OCR

Cohere Aya Vision 32B 5 / 9

Gemma 3 4B 6 / 9

Object Counting

Cohere Aya Vision 32B 2 / 10

Gemma 3 4B 0 / 10

Object Detection

Cohere Aya Vision 32B 2 / 2

Gemma 3 4B 2 / 2

Object Measurement

Cohere Aya Vision 32B 0 / 2

Gemma 3 4B 0 / 2

Object Understanding

Cohere Aya Vision 32B 8 / 11

Gemma 3 4B 5 / 11

Receipt Reading

Cohere Aya Vision 32B 0 / 1

Gemma 3 4B 0 / 1

Sign Understanding

Cohere Aya Vision 32B 2 / 2

Gemma 3 4B 2 / 2

Spatial Relations

Cohere Aya Vision 32B 10 / 17

Gemma 3 4B 4 / 17

Web Action Understanding

Cohere Aya Vision 32B 4 / 4

Gemma 3 4B 2 / 4

Contribute a Prompt

Lightbulb

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