GPT-5.4 vs.Qwen 3.5 9B
Passed:
GPT-5.4 79.3% (65/82)
Qwen 3.5 9B 67.1% (55/82)
Average request time:
GPT-5.4 9.40s
Qwen 3.5 9B 7.99s
Passed:
GPT-5.4 79.3% (65/82)
Qwen 3.5 9B 67.1% (55/82)
Average request time:
GPT-5.4 9.40s
Qwen 3.5 9B 7.99s
Summary
Annotation Understanding
GPT-5.4
2 / 2
Qwen 3.5 9B
0 / 2
CAPTCHA
GPT-5.4
0 / 2
Qwen 3.5 9B
2 / 2
Color Identification
GPT-5.4
1 / 1
Qwen 3.5 9B
1 / 1
Defect Detection
GPT-5.4
13 / 15
Qwen 3.5 9B
10 / 15
Document Understanding
GPT-5.4
9 / 10
Qwen 3.5 9B
5 / 10
Localization
GPT-5.4
1 / 1
Qwen 3.5 9B
0 / 1
OCR
GPT-5.4
9 / 9
Qwen 3.5 9B
6 / 9
Object Counting
GPT-5.4
4 / 11
Qwen 3.5 9B
4 / 11
Object Detection
GPT-5.4
2 / 2
Qwen 3.5 9B
2 / 2
Object Measurement
GPT-5.4
0 / 1
Qwen 3.5 9B
0 / 1
Object Understanding
GPT-5.4
8 / 11
Qwen 3.5 9B
9 / 11
Receipt Reading
GPT-5.4
1 / 1
Qwen 3.5 9B
1 / 1
Sign Understanding
GPT-5.4
2 / 2
Qwen 3.5 9B
2 / 2
Spatial Relations
GPT-5.4
15 / 17
Qwen 3.5 9B
16 / 17
Web Action Understanding
GPT-5.4
4 / 4
Qwen 3.5 9B
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!