Gemma 3 27b vs.Cosmos Reason2 2B
Passed:
Gemma 3 27b 63.0% (51/81)
Cosmos Reason2 2B 41.5% (34/82)
Average request time:
Gemma 3 27b 30.87s
Cosmos Reason2 2B 1.94s
Passed:
Gemma 3 27b 63.0% (51/81)
Cosmos Reason2 2B 41.5% (34/82)
Average request time:
Gemma 3 27b 30.87s
Cosmos Reason2 2B 1.94s
Summary
Annotation Understanding
Gemma 3 27b
0 / 2
Cosmos Reason2 2B
0 / 2
CAPTCHA
Gemma 3 27b
2 / 2
Cosmos Reason2 2B
2 / 2
Color Identification
Gemma 3 27b
0 / 1
Cosmos Reason2 2B
1 / 1
Defect Detection
Gemma 3 27b
9 / 15
Cosmos Reason2 2B
9 / 15
Document Understanding
Gemma 3 27b
8 / 10
Cosmos Reason2 2B
4 / 10
Localization
Gemma 3 27b
1 / 1
Cosmos Reason2 2B
0 / 1
OCR
Gemma 3 27b
6 / 9
Cosmos Reason2 2B
6 / 9
Object Counting
Gemma 3 27b
1 / 9
Cosmos Reason2 2B
1 / 11
Object Detection
Gemma 3 27b
2 / 2
Cosmos Reason2 2B
1 / 2
Object Measurement
Gemma 3 27b
0 / 1
Cosmos Reason2 2B
0 / 1
Object Understanding
Gemma 3 27b
8 / 11
Cosmos Reason2 2B
6 / 11
Receipt Reading
Gemma 3 27b
1 / 1
Cosmos Reason2 2B
1 / 1
Sign Understanding
Gemma 3 27b
2 / 2
Cosmos Reason2 2B
2 / 2
Spatial Relations
Gemma 3 27b
11 / 17
Cosmos Reason2 2B
4 / 17
Web Action Understanding
Gemma 3 27b
4 / 4
Cosmos Reason2 2B
0 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!