Qwen 2.5 VL 7B vs.Claude 4 Opus
Passed:
Qwen 2.5 VL 7B 55.6% (45/81)
Claude 4 Opus 62.7% (52/83)
Average request time:
Qwen 2.5 VL 7B 45.63s
Claude 4 Opus 17.98s
Passed:
Qwen 2.5 VL 7B 55.6% (45/81)
Claude 4 Opus 62.7% (52/83)
Average request time:
Qwen 2.5 VL 7B 45.63s
Claude 4 Opus 17.98s
Summary
Annotation Understanding
Qwen 2.5 VL 7B
1 / 2
Claude 4 Opus
1 / 2
CAPTCHA
Qwen 2.5 VL 7B
2 / 2
Claude 4 Opus
2 / 2
Color Identification
Qwen 2.5 VL 7B
0 / 1
Claude 4 Opus
1 / 1
Counting nodes
Qwen 2.5 VL 7B
0 / 0
Claude 4 Opus
0 / 1
Defect Detection
Qwen 2.5 VL 7B
9 / 15
Claude 4 Opus
10 / 15
Document Understanding
Qwen 2.5 VL 7B
8 / 10
Claude 4 Opus
9 / 10
Localization
Qwen 2.5 VL 7B
1 / 1
Claude 4 Opus
0 / 1
OCR
Qwen 2.5 VL 7B
6 / 9
Claude 4 Opus
8 / 9
Object Counting
Qwen 2.5 VL 7B
1 / 9
Claude 4 Opus
0 / 10
Object Detection
Qwen 2.5 VL 7B
1 / 2
Claude 4 Opus
2 / 2
Object Measurement
Qwen 2.5 VL 7B
0 / 2
Claude 4 Opus
1 / 2
Object Understanding
Qwen 2.5 VL 7B
5 / 11
Claude 4 Opus
7 / 11
Receipt Reading
Qwen 2.5 VL 7B
0 / 1
Claude 4 Opus
1 / 1
Sign Understanding
Qwen 2.5 VL 7B
2 / 2
Claude 4 Opus
1 / 2
Spatial Relations
Qwen 2.5 VL 7B
11 / 17
Claude 4 Opus
9 / 17
Web Action Understanding
Qwen 2.5 VL 7B
4 / 4
Claude 4 Opus
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!