Claude 3.7 Sonnet vs.Cohere Aya Vision 32B
Passed:
Claude 3.7 Sonnet 63.9% (53/83)
Cohere Aya Vision 32B 56.6% (47/83)
Average request time:
Claude 3.7 Sonnet 13.98s
Cohere Aya Vision 32B 18.49s
Passed:
Claude 3.7 Sonnet 63.9% (53/83)
Cohere Aya Vision 32B 56.6% (47/83)
Average request time:
Claude 3.7 Sonnet 13.98s
Cohere Aya Vision 32B 18.49s
Summary
Annotation Understanding
Claude 3.7 Sonnet
0 / 2
Cohere Aya Vision 32B
0 / 2
CAPTCHA
Claude 3.7 Sonnet
2 / 2
Cohere Aya Vision 32B
0 / 2
Color Identification
Claude 3.7 Sonnet
1 / 1
Cohere Aya Vision 32B
0 / 1
Counting nodes
Claude 3.7 Sonnet
0 / 1
Cohere Aya Vision 32B
0 / 1
Defect Detection
Claude 3.7 Sonnet
10 / 15
Cohere Aya Vision 32B
10 / 15
Document Understanding
Claude 3.7 Sonnet
9 / 10
Cohere Aya Vision 32B
6 / 10
Localization
Claude 3.7 Sonnet
1 / 1
Cohere Aya Vision 32B
1 / 1
OCR
Claude 3.7 Sonnet
7 / 9
Cohere Aya Vision 32B
5 / 9
Object Counting
Claude 3.7 Sonnet
2 / 10
Cohere Aya Vision 32B
2 / 10
Object Detection
Claude 3.7 Sonnet
2 / 2
Cohere Aya Vision 32B
2 / 2
Object Measurement
Claude 3.7 Sonnet
0 / 2
Cohere Aya Vision 32B
0 / 2
Object Understanding
Claude 3.7 Sonnet
7 / 11
Cohere Aya Vision 32B
8 / 11
Receipt Reading
Claude 3.7 Sonnet
1 / 1
Cohere Aya Vision 32B
0 / 1
Sign Understanding
Claude 3.7 Sonnet
2 / 2
Cohere Aya Vision 32B
2 / 2
Spatial Relations
Claude 3.7 Sonnet
9 / 17
Cohere Aya Vision 32B
10 / 17
Web Action Understanding
Claude 3.7 Sonnet
4 / 4
Cohere Aya Vision 32B
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!