adjustable glasses illustration eye chart illustration clipboard illustration lenses illustration

Phi 4 Multimodal vs.Qwen 3.5 122B (A10B)

Passed:
Phi 4 Multimodal 49.4% (41/83)
Qwen 3.5 122B (A10B) 78.0% (64/82)
Average request time:
Phi 4 Multimodal 15.92s
Qwen 3.5 122B (A10B) 1.54s

Summary

Annotation Understanding
Phi 4 Multimodal 1 / 2
Qwen 3.5 122B (A10B) 1 / 2
CAPTCHA
Phi 4 Multimodal 0 / 2
Qwen 3.5 122B (A10B) 2 / 2
Color Identification
Phi 4 Multimodal 0 / 1
Qwen 3.5 122B (A10B) 1 / 1
Defect Detection
Phi 4 Multimodal 11 / 15
Qwen 3.5 122B (A10B) 13 / 15
Document Understanding
Phi 4 Multimodal 7 / 10
Qwen 3.5 122B (A10B) 9 / 10
Localization
Phi 4 Multimodal 0 / 1
Qwen 3.5 122B (A10B) 1 / 1
OCR
Phi 4 Multimodal 5 / 9
Qwen 3.5 122B (A10B) 7 / 9
Object Counting
Phi 4 Multimodal 0 / 11
Qwen 3.5 122B (A10B) 5 / 11
Object Detection
Phi 4 Multimodal 2 / 2
Qwen 3.5 122B (A10B) 2 / 2
Object Measurement
Phi 4 Multimodal 0 / 1
Qwen 3.5 122B (A10B) 1 / 1
Object Understanding
Phi 4 Multimodal 5 / 11
Qwen 3.5 122B (A10B) 8 / 11
Receipt Reading
Phi 4 Multimodal 1 / 1
Qwen 3.5 122B (A10B) 1 / 1
Sign Understanding
Phi 4 Multimodal 2 / 2
Qwen 3.5 122B (A10B) 2 / 2
Spatial Relations
Phi 4 Multimodal 7 / 17
Qwen 3.5 122B (A10B) 14 / 17
Web Action Understanding
Phi 4 Multimodal 2 / 4
Qwen 3.5 122B (A10B) 4 / 4

Contribute a Prompt