GPT-4.1 vs.Gemma 3 4B
Passed:
GPT-4.1 73.2% (60/82)
Gemma 3 4B 45.1% (37/82)
Average request time:
GPT-4.1 14.91s
Gemma 3 4B 2.66s
Passed:
GPT-4.1 73.2% (60/82)
Gemma 3 4B 45.1% (37/82)
Average request time:
GPT-4.1 14.91s
Gemma 3 4B 2.66s
Summary
Annotation Understanding
GPT-4.1
1 / 2
Gemma 3 4B
1 / 2
CAPTCHA
GPT-4.1
0 / 2
Gemma 3 4B
2 / 2
Color Identification
GPT-4.1
1 / 1
Gemma 3 4B
1 / 1
Counting nodes
GPT-4.1
0 / 1
Gemma 3 4B
0 / 1
Defect Detection
GPT-4.1
13 / 15
Gemma 3 4B
9 / 15
Document Understanding
GPT-4.1
5 / 10
Gemma 3 4B
4 / 10
Localization
GPT-4.1
1 / 1
Gemma 3 4B
0 / 1
OCR
GPT-4.1
9 / 9
Gemma 3 4B
5 / 9
Object Counting
GPT-4.1
2 / 10
Gemma 3 4B
1 / 10
Object Detection
GPT-4.1
2 / 2
Gemma 3 4B
2 / 2
Object Measurement
GPT-4.1
1 / 2
Gemma 3 4B
0 / 2
Object Understanding
GPT-4.1
9 / 10
Gemma 3 4B
5 / 10
Receipt Reading
GPT-4.1
1 / 1
Gemma 3 4B
0 / 1
Sign Understanding
GPT-4.1
2 / 2
Gemma 3 4B
2 / 2
Spatial Relations
GPT-4.1
14 / 17
Gemma 3 4B
8 / 17
Web Action Understanding
GPT-4.1
4 / 4
Gemma 3 4B
3 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!