Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Desktop UI Navigation on OSWorld 50 easy tasks 1.0 (test)
Loading...
100
ASR
Claude 3.5 Sonnet
79.2
84.6
90
95.4
Nov 4, 2024
ASR
SR
OSR
Updated 1mo ago
Evaluation Results
Method
Method
Links
ASR
SR
OSR
Claude 3.5 Sonnet
Observation Modality=S...
2024.11
100
0
22
Claude 3.5 Sonnet v2
Observation Modality=S...
2024.11
96
4
22
GPT-4o
Observation Modality=S...
2024.11
95.8
6
8
Claude 3.5 Sonnet
Observation Modality=S...
2024.11
95.3
6
44
Claude 3.5 Sonnet v2
Observation Modality=S...
2024.11
94.8
10
58
GPT-4-Turbo
Observation Modality=S...
2024.11
93.3
2
18
GPT-4-Turbo
Observation Modality=S...
2024.11
91.8
8
52
GPT-4o
Observation Modality=S...
2024.11
91.2
2
6
Gemini 1.5
Observation Modality=S...
2024.11
88.7
6
18
Gemini 1.5
Observation Modality=S...
2024.11
80
4
6
Feedback
Search any
task
Search any
task