Gemma 3 27b vs.GLM 4.6v
Passed:
Gemma 3 27b 63.0% (51/81)
GLM 4.6v 68.0% (51/75)
Average request time:
Gemma 3 27b 30.87s
GLM 4.6v 12.37s
Passed:
Gemma 3 27b 63.0% (51/81)
GLM 4.6v 68.0% (51/75)
Average request time:
Gemma 3 27b 30.87s
GLM 4.6v 12.37s
Summary
Annotation Understanding
Gemma 3 27b
0 / 2
GLM 4.6v
0 / 0
CAPTCHA
Gemma 3 27b
2 / 2
GLM 4.6v
2 / 2
Color Identification
Gemma 3 27b
0 / 1
GLM 4.6v
0 / 1
Defect Detection
Gemma 3 27b
9 / 15
GLM 4.6v
11 / 15
Document Understanding
Gemma 3 27b
8 / 10
GLM 4.6v
7 / 8
Localization
Gemma 3 27b
1 / 1
GLM 4.6v
1 / 1
OCR
Gemma 3 27b
6 / 9
GLM 4.6v
7 / 9
Object Counting
Gemma 3 27b
1 / 9
GLM 4.6v
2 / 9
Object Detection
Gemma 3 27b
2 / 2
GLM 4.6v
0 / 2
Object Measurement
Gemma 3 27b
0 / 1
GLM 4.6v
0 / 1
Object Understanding
Gemma 3 27b
8 / 11
GLM 4.6v
8 / 10
Receipt Reading
Gemma 3 27b
1 / 1
GLM 4.6v
1 / 1
Sign Understanding
Gemma 3 27b
2 / 2
GLM 4.6v
2 / 2
Spatial Relations
Gemma 3 27b
11 / 17
GLM 4.6v
9 / 17
Web Action Understanding
Gemma 3 27b
4 / 4
GLM 4.6v
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!