Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
UI Task Completion on AITW
Loading...
7.7
Overall Task Time
ChatGPT-COT
5.7
19.2
32.7
46.2
Aug 1, 2024
Overall Task Time
General Task Time
Install Task Time
Google Apps Task Time
Single Task Time
Web Shopping Task Time
Updated 1mo ago
Evaluation Results
Method
Method
Links
Overall Task Time
General Task Time
Install Task Time
Google Apps Task Time
Single Task Time
Web Shopping Task Time
ChatGPT-COT
Modality=Text
2024.08
7.7
5.9
4.4
10.5
9.4
8.4
PaLM2-COT
Modality=Text
2024.08
39.6
-
-
-
-
-
GPT-4V image-only
Modality=Image
2024.08
50.5
41.7
42.6
49.8
72.8
45.7
GPT-4V + history
Modality=Image
2024.08
53
43
46.1
49.2
78.3
48.2
OmniParser (w. LS + ID)
Modality=Image
2024.08
57.7
48.3
57.8
51.6
77.4
52.9
Feedback
Search any
task
Search any
task