OpenAI O1 vs.OpenAI O4 Mini (Medium Reasoning)
Passed:
OpenAI O1 65.9% (54/82)
OpenAI O4 Mini (Medium Reasoning) 79.3% (65/82)
Average request time:
OpenAI O1 29.07s
OpenAI O4 Mini (Medium Reasoning) 12.87s
Passed:
OpenAI O1 65.9% (54/82)
OpenAI O4 Mini (Medium Reasoning) 79.3% (65/82)
Average request time:
OpenAI O1 29.07s
OpenAI O4 Mini (Medium Reasoning) 12.87s
Summary
Annotation Understanding
OpenAI O1
1 / 2
OpenAI O4 Mini (Medium Reasoning)
2 / 2
CAPTCHA
OpenAI O1
2 / 2
OpenAI O4 Mini (Medium Reasoning)
2 / 2
Color Identification
OpenAI O1
0 / 1
OpenAI O4 Mini (Medium Reasoning)
1 / 1
Defect Detection
OpenAI O1
11 / 15
OpenAI O4 Mini (Medium Reasoning)
14 / 15
Document Understanding
OpenAI O1
7 / 10
OpenAI O4 Mini (Medium Reasoning)
7 / 10
Localization
OpenAI O1
1 / 1
OpenAI O4 Mini (Medium Reasoning)
1 / 1
OCR
OpenAI O1
9 / 9
OpenAI O4 Mini (Medium Reasoning)
9 / 9
Object Counting
OpenAI O1
2 / 11
OpenAI O4 Mini (Medium Reasoning)
3 / 11
Object Detection
OpenAI O1
1 / 2
OpenAI O4 Mini (Medium Reasoning)
1 / 2
Object Measurement
OpenAI O1
1 / 1
OpenAI O4 Mini (Medium Reasoning)
0 / 1
Object Understanding
OpenAI O1
5 / 11
OpenAI O4 Mini (Medium Reasoning)
10 / 11
Receipt Reading
OpenAI O1
0 / 1
OpenAI O4 Mini (Medium Reasoning)
1 / 1
Sign Understanding
OpenAI O1
2 / 2
OpenAI O4 Mini (Medium Reasoning)
2 / 2
Spatial Relations
OpenAI O1
14 / 17
OpenAI O4 Mini (Medium Reasoning)
15 / 17
Web Action Understanding
OpenAI O1
4 / 4
OpenAI O4 Mini (Medium Reasoning)
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!