Claude 3.5 Haiku vs.Qwen 3.6 Plus
Passed:
Claude 3.5 Haiku 65.1% (54/83)
Qwen 3.6 Plus 74.4% (61/82)
Average request time:
Claude 3.5 Haiku 16.75s
Qwen 3.6 Plus 81.82s
Passed:
Claude 3.5 Haiku 65.1% (54/83)
Qwen 3.6 Plus 74.4% (61/82)
Average request time:
Claude 3.5 Haiku 16.75s
Qwen 3.6 Plus 81.82s
Summary
Annotation Understanding
Claude 3.5 Haiku
2 / 2
Qwen 3.6 Plus
1 / 2
CAPTCHA
Claude 3.5 Haiku
0 / 2
Qwen 3.6 Plus
2 / 2
Color Identification
Claude 3.5 Haiku
0 / 1
Qwen 3.6 Plus
1 / 1
Defect Detection
Claude 3.5 Haiku
10 / 15
Qwen 3.6 Plus
11 / 15
Document Understanding
Claude 3.5 Haiku
8 / 10
Qwen 3.6 Plus
9 / 10
Localization
Claude 3.5 Haiku
0 / 1
Qwen 3.6 Plus
0 / 1
OCR
Claude 3.5 Haiku
7 / 9
Qwen 3.6 Plus
7 / 9
Object Counting
Claude 3.5 Haiku
5 / 11
Qwen 3.6 Plus
4 / 11
Object Detection
Claude 3.5 Haiku
2 / 2
Qwen 3.6 Plus
1 / 2
Object Measurement
Claude 3.5 Haiku
0 / 1
Qwen 3.6 Plus
1 / 1
Object Understanding
Claude 3.5 Haiku
7 / 11
Qwen 3.6 Plus
10 / 11
Receipt Reading
Claude 3.5 Haiku
1 / 1
Qwen 3.6 Plus
1 / 1
Sign Understanding
Claude 3.5 Haiku
2 / 2
Qwen 3.6 Plus
2 / 2
Spatial Relations
Claude 3.5 Haiku
8 / 17
Qwen 3.6 Plus
14 / 17
Web Action Understanding
Claude 3.5 Haiku
4 / 4
Qwen 3.6 Plus
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!