Gemini 2.5 Pro vs.Qwen 2.5 VL 7B
Passed:
Gemini 2.5 Pro 74.4% (61/82)
Qwen 2.5 VL 7B 55.6% (45/81)
Average request time:
Gemini 2.5 Pro 9.95s
Qwen 2.5 VL 7B 45.63s
Passed:
Gemini 2.5 Pro 74.4% (61/82)
Qwen 2.5 VL 7B 55.6% (45/81)
Average request time:
Gemini 2.5 Pro 9.95s
Qwen 2.5 VL 7B 45.63s
Summary
Annotation Understanding
Gemini 2.5 Pro
1 / 2
Qwen 2.5 VL 7B
1 / 2
CAPTCHA
Gemini 2.5 Pro
2 / 2
Qwen 2.5 VL 7B
2 / 2
Color Identification
Gemini 2.5 Pro
0 / 1
Qwen 2.5 VL 7B
0 / 1
Defect Detection
Gemini 2.5 Pro
10 / 15
Qwen 2.5 VL 7B
9 / 15
Document Understanding
Gemini 2.5 Pro
9 / 10
Qwen 2.5 VL 7B
8 / 10
Localization
Gemini 2.5 Pro
1 / 1
Qwen 2.5 VL 7B
1 / 1
OCR
Gemini 2.5 Pro
9 / 9
Qwen 2.5 VL 7B
6 / 9
Object Counting
Gemini 2.5 Pro
4 / 11
Qwen 2.5 VL 7B
1 / 9
Object Detection
Gemini 2.5 Pro
2 / 2
Qwen 2.5 VL 7B
1 / 2
Object Measurement
Gemini 2.5 Pro
0 / 1
Qwen 2.5 VL 7B
0 / 1
Object Understanding
Gemini 2.5 Pro
8 / 11
Qwen 2.5 VL 7B
5 / 11
Receipt Reading
Gemini 2.5 Pro
1 / 1
Qwen 2.5 VL 7B
0 / 1
Sign Understanding
Gemini 2.5 Pro
2 / 2
Qwen 2.5 VL 7B
2 / 2
Spatial Relations
Gemini 2.5 Pro
14 / 17
Qwen 2.5 VL 7B
11 / 17
Web Action Understanding
Gemini 2.5 Pro
4 / 4
Qwen 2.5 VL 7B
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!