Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
OS-level Agent Task Execution on OSWorld subset
Loading...
46.48
Success Rate
EVA-CUA-8B
22.5496
28.7623
34.975
41.1877
Jun 10, 2026
Success Rate
Average Steps
Teacher Queries per Task
Updated 1mo ago
Evaluation Results
Method
Method
Links
Success Rate
Average Steps
Teacher Queries per Task
EVA-CUA-8B
Deployment Mode=teache...
2026.06
46.48
-
-
SFT collector with SRC
Deployment Mode=teache...
2026.06
43.27
16.82
5.29
Expert SFT + SRC
Deployment Mode=teache...
2026.06
40.15
11.86
-
Expert SFT
Deployment Mode=teache...
2026.06
27.22
12.81
-
Base model
Deployment Mode=teache...
2026.06
23.47
12.88
-
Feedback
Search any
task
Search any
task