Phi 4 Multimodal vs.Cosmos Reason2 8B
Passed:
Phi 4 Multimodal 49.4% (41/83)
Cosmos Reason2 8B 63.4% (52/82)
Average request time:
Phi 4 Multimodal 15.92s
Cosmos Reason2 8B 2.72s
Passed:
Phi 4 Multimodal 49.4% (41/83)
Cosmos Reason2 8B 63.4% (52/82)
Average request time:
Phi 4 Multimodal 15.92s
Cosmos Reason2 8B 2.72s
Summary
Annotation Understanding
Phi 4 Multimodal
1 / 2
Cosmos Reason2 8B
2 / 2
CAPTCHA
Phi 4 Multimodal
0 / 2
Cosmos Reason2 8B
2 / 2
Color Identification
Phi 4 Multimodal
0 / 1
Cosmos Reason2 8B
1 / 1
Defect Detection
Phi 4 Multimodal
11 / 15
Cosmos Reason2 8B
10 / 15
Document Understanding
Phi 4 Multimodal
7 / 10
Cosmos Reason2 8B
5 / 10
Localization
Phi 4 Multimodal
0 / 1
Cosmos Reason2 8B
1 / 1
OCR
Phi 4 Multimodal
5 / 9
Cosmos Reason2 8B
8 / 9
Object Counting
Phi 4 Multimodal
0 / 11
Cosmos Reason2 8B
3 / 11
Object Detection
Phi 4 Multimodal
2 / 2
Cosmos Reason2 8B
0 / 2
Object Measurement
Phi 4 Multimodal
0 / 1
Cosmos Reason2 8B
0 / 1
Object Understanding
Phi 4 Multimodal
5 / 11
Cosmos Reason2 8B
8 / 11
Receipt Reading
Phi 4 Multimodal
1 / 1
Cosmos Reason2 8B
1 / 1
Sign Understanding
Phi 4 Multimodal
2 / 2
Cosmos Reason2 8B
2 / 2
Spatial Relations
Phi 4 Multimodal
7 / 17
Cosmos Reason2 8B
10 / 17
Web Action Understanding
Phi 4 Multimodal
2 / 4
Cosmos Reason2 8B
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!