adjustable glasses illustration eye chart illustration clipboard illustration lenses illustration

Gemma 3 4B vs.Phi 4 Multimodal

Passed:
Gemma 3 4B 45.1% (37/82)
Phi 4 Multimodal 56.1% (46/82)
Average request time:
Gemma 3 4B 2.66s
Phi 4 Multimodal 4.08s

Summary

Annotation Understanding
Gemma 3 4B 1 / 2
Phi 4 Multimodal 2 / 2
CAPTCHA
Gemma 3 4B 2 / 2
Phi 4 Multimodal 0 / 2
Color Identification
Gemma 3 4B 1 / 1
Phi 4 Multimodal 1 / 1
Counting nodes
Gemma 3 4B 0 / 1
Phi 4 Multimodal 0 / 1
Defect Detection
Gemma 3 4B 9 / 15
Phi 4 Multimodal 12 / 15
Document Understanding
Gemma 3 4B 4 / 10
Phi 4 Multimodal 4 / 10
Localization
Gemma 3 4B 0 / 1
Phi 4 Multimodal 0 / 1
OCR
Gemma 3 4B 5 / 9
Phi 4 Multimodal 6 / 9
Object Counting
Gemma 3 4B 1 / 10
Phi 4 Multimodal 0 / 10
Object Detection
Gemma 3 4B 2 / 2
Phi 4 Multimodal 1 / 2
Object Measurement
Gemma 3 4B 0 / 2
Phi 4 Multimodal 0 / 2
Object Understanding
Gemma 3 4B 5 / 10
Phi 4 Multimodal 8 / 10
Receipt Reading
Gemma 3 4B 0 / 1
Phi 4 Multimodal 1 / 1
Sign Understanding
Gemma 3 4B 2 / 2
Phi 4 Multimodal 1 / 2
Spatial Relations
Gemma 3 4B 8 / 17
Phi 4 Multimodal 9 / 17
Web Action Understanding
Gemma 3 4B 3 / 4
Phi 4 Multimodal 3 / 4

Contribute a Prompt