Llama 4 Scout 17B vs.Qwen 3.5 122B (A10B)
Passed:
Llama 4 Scout 17B 72.8% (59/81)
Qwen 3.5 122B (A10B) 78.0% (64/82)
Average request time:
Llama 4 Scout 17B 41.93s
Qwen 3.5 122B (A10B) 1.54s
Passed:
Llama 4 Scout 17B 72.8% (59/81)
Qwen 3.5 122B (A10B) 78.0% (64/82)
Average request time:
Llama 4 Scout 17B 41.93s
Qwen 3.5 122B (A10B) 1.54s
Summary
Annotation Understanding
Llama 4 Scout 17B
0 / 2
Qwen 3.5 122B (A10B)
1 / 2
CAPTCHA
Llama 4 Scout 17B
2 / 2
Qwen 3.5 122B (A10B)
2 / 2
Color Identification
Llama 4 Scout 17B
0 / 1
Qwen 3.5 122B (A10B)
1 / 1
Defect Detection
Llama 4 Scout 17B
11 / 15
Qwen 3.5 122B (A10B)
13 / 15
Document Understanding
Llama 4 Scout 17B
9 / 10
Qwen 3.5 122B (A10B)
9 / 10
Localization
Llama 4 Scout 17B
1 / 1
Qwen 3.5 122B (A10B)
1 / 1
OCR
Llama 4 Scout 17B
9 / 9
Qwen 3.5 122B (A10B)
7 / 9
Object Counting
Llama 4 Scout 17B
2 / 9
Qwen 3.5 122B (A10B)
5 / 11
Object Detection
Llama 4 Scout 17B
2 / 2
Qwen 3.5 122B (A10B)
2 / 2
Object Measurement
Llama 4 Scout 17B
0 / 1
Qwen 3.5 122B (A10B)
1 / 1
Object Understanding
Llama 4 Scout 17B
9 / 11
Qwen 3.5 122B (A10B)
8 / 11
Receipt Reading
Llama 4 Scout 17B
0 / 1
Qwen 3.5 122B (A10B)
1 / 1
Sign Understanding
Llama 4 Scout 17B
2 / 2
Qwen 3.5 122B (A10B)
2 / 2
Spatial Relations
Llama 4 Scout 17B
12 / 17
Qwen 3.5 122B (A10B)
14 / 17
Web Action Understanding
Llama 4 Scout 17B
4 / 4
Qwen 3.5 122B (A10B)
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!