GPT-5.4 vs.OpenAI O1
Passed:
GPT-5.4 79.3% (65/82)
OpenAI O1 65.9% (54/82)
Average request time:
GPT-5.4 9.40s
OpenAI O1 29.07s
Passed:
GPT-5.4 79.3% (65/82)
OpenAI O1 65.9% (54/82)
Average request time:
GPT-5.4 9.40s
OpenAI O1 29.07s
Summary
Annotation Understanding
GPT-5.4
2 / 2
OpenAI O1
1 / 2
CAPTCHA
GPT-5.4
0 / 2
OpenAI O1
2 / 2
Color Identification
GPT-5.4
1 / 1
OpenAI O1
0 / 1
Defect Detection
GPT-5.4
13 / 15
OpenAI O1
11 / 15
Document Understanding
GPT-5.4
9 / 10
OpenAI O1
7 / 10
Localization
GPT-5.4
1 / 1
OpenAI O1
1 / 1
OCR
GPT-5.4
9 / 9
OpenAI O1
9 / 9
Object Counting
GPT-5.4
4 / 11
OpenAI O1
2 / 11
Object Detection
GPT-5.4
2 / 2
OpenAI O1
1 / 2
Object Measurement
GPT-5.4
0 / 1
OpenAI O1
1 / 1
Object Understanding
GPT-5.4
8 / 11
OpenAI O1
5 / 11
Receipt Reading
GPT-5.4
1 / 1
OpenAI O1
0 / 1
Sign Understanding
GPT-5.4
2 / 2
OpenAI O1
2 / 2
Spatial Relations
GPT-5.4
15 / 17
OpenAI O1
14 / 17
Web Action Understanding
GPT-5.4
4 / 4
OpenAI O1
4 / 4
Contribute a Prompt
Have an idea for a prompt? Open a pull request on the project repository!