Phi 4 Multimodal vs.GPT-5.2
Passed:
Phi 4 Multimodal 49.4% (41/83)
GPT-5.2 79.3% (65/82)
Average request time:
Phi 4 Multimodal 15.92s
GPT-5.2 10.81s
Passed:
Phi 4 Multimodal 49.4% (41/83)
GPT-5.2 79.3% (65/82)
Average request time:
Phi 4 Multimodal 15.92s
GPT-5.2 10.81s
Summary
Annotation Understanding
Phi 4 Multimodal
1 / 2
GPT-5.2
1 / 2
CAPTCHA
Phi 4 Multimodal
0 / 2
GPT-5.2
0 / 2
Color Identification
Phi 4 Multimodal
0 / 1
GPT-5.2
1 / 1
Defect Detection
Phi 4 Multimodal
11 / 15
GPT-5.2
13 / 15
Document Understanding
Phi 4 Multimodal
7 / 10
GPT-5.2
9 / 10
Localization
Phi 4 Multimodal
0 / 1
GPT-5.2
1 / 1
OCR
Phi 4 Multimodal
5 / 9
GPT-5.2
9 / 9
Object Counting
Phi 4 Multimodal
0 / 11
GPT-5.2
4 / 11
Object Detection
Phi 4 Multimodal
2 / 2
GPT-5.2
1 / 2
Object Measurement
Phi 4 Multimodal
0 / 1
GPT-5.2
0 / 1
Object Understanding
Phi 4 Multimodal
5 / 11
GPT-5.2
10 / 11
Receipt Reading
Phi 4 Multimodal
1 / 1
GPT-5.2
1 / 1
Sign Understanding
Phi 4 Multimodal
2 / 2
GPT-5.2
2 / 2
Spatial Relations
Phi 4 Multimodal
7 / 17
GPT-5.2
15 / 17
Web Action Understanding
Phi 4 Multimodal
2 / 4
GPT-5.2
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!