Qwen 2.5 VL 7B vs.Mistral Medium 3
Passed:
Qwen 2.5 VL 7B 55.6% (45/81)
Mistral Medium 3 68.7% (57/83)
Average request time:
Qwen 2.5 VL 7B 45.63s
Mistral Medium 3 13.13s
Passed:
Qwen 2.5 VL 7B 55.6% (45/81)
Mistral Medium 3 68.7% (57/83)
Average request time:
Qwen 2.5 VL 7B 45.63s
Mistral Medium 3 13.13s
Summary
Annotation Understanding
Qwen 2.5 VL 7B
1 / 2
Mistral Medium 3
1 / 2
CAPTCHA
Qwen 2.5 VL 7B
2 / 2
Mistral Medium 3
2 / 2
Color Identification
Qwen 2.5 VL 7B
0 / 1
Mistral Medium 3
0 / 1
Counting nodes
Qwen 2.5 VL 7B
0 / 0
Mistral Medium 3
0 / 1
Defect Detection
Qwen 2.5 VL 7B
9 / 15
Mistral Medium 3
12 / 15
Document Understanding
Qwen 2.5 VL 7B
8 / 10
Mistral Medium 3
8 / 10
Localization
Qwen 2.5 VL 7B
1 / 1
Mistral Medium 3
1 / 1
OCR
Qwen 2.5 VL 7B
6 / 9
Mistral Medium 3
9 / 9
Object Counting
Qwen 2.5 VL 7B
1 / 9
Mistral Medium 3
2 / 10
Object Detection
Qwen 2.5 VL 7B
1 / 2
Mistral Medium 3
2 / 2
Object Measurement
Qwen 2.5 VL 7B
0 / 2
Mistral Medium 3
0 / 2
Object Understanding
Qwen 2.5 VL 7B
5 / 11
Mistral Medium 3
9 / 11
Receipt Reading
Qwen 2.5 VL 7B
0 / 1
Mistral Medium 3
1 / 1
Sign Understanding
Qwen 2.5 VL 7B
2 / 2
Mistral Medium 3
2 / 2
Spatial Relations
Qwen 2.5 VL 7B
11 / 17
Mistral Medium 3
7 / 17
Web Action Understanding
Qwen 2.5 VL 7B
4 / 4
Mistral Medium 3
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!