Kimi k2.5 vs.Qwen 3.5 27B
Passed:
Kimi k2.5 73.5% (50/68)
Qwen 3.5 27B 74.4% (61/82)
Average request time:
Kimi k2.5 44.70s
Qwen 3.5 27B 1.89s
Passed:
Kimi k2.5 73.5% (50/68)
Qwen 3.5 27B 74.4% (61/82)
Average request time:
Kimi k2.5 44.70s
Qwen 3.5 27B 1.89s
Summary
Annotation Understanding
Kimi k2.5
1 / 1
Qwen 3.5 27B
1 / 2
CAPTCHA
Kimi k2.5
2 / 2
Qwen 3.5 27B
2 / 2
Color Identification
Kimi k2.5
1 / 1
Qwen 3.5 27B
1 / 1
Defect Detection
Kimi k2.5
8 / 14
Qwen 3.5 27B
12 / 15
Document Understanding
Kimi k2.5
8 / 8
Qwen 3.5 27B
7 / 10
Localization
Kimi k2.5
1 / 1
Qwen 3.5 27B
1 / 1
OCR
Kimi k2.5
7 / 9
Qwen 3.5 27B
8 / 9
Object Counting
Kimi k2.5
1 / 3
Qwen 3.5 27B
5 / 11
Object Detection
Kimi k2.5
1 / 2
Qwen 3.5 27B
1 / 2
Object Measurement
Kimi k2.5
0 / 1
Qwen 3.5 27B
1 / 1
Object Understanding
Kimi k2.5
9 / 11
Qwen 3.5 27B
9 / 11
Receipt Reading
Kimi k2.5
1 / 1
Qwen 3.5 27B
1 / 1
Sign Understanding
Kimi k2.5
2 / 2
Qwen 3.5 27B
2 / 2
Spatial Relations
Kimi k2.5
12 / 15
Qwen 3.5 27B
13 / 17
Web Action Understanding
Kimi k2.5
3 / 4
Qwen 3.5 27B
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!