Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Strategic Gaming on Civilization V
Loading...
0
Turn Count
CRADLE
-4.2932
24.6859
53.665
82.6441
Oct 17, 2025
Turn Count
Technology Progress
Error Rate (ERR)
Cost
Updated 24d ago
Evaluation Results
Method
Method
Links
Turn Count
Technology Progress
Error Rate (ERR)
Cost
CRADLE
Setting=prior-assisted
2025.10
0
0
-
4.23
Claude3.7
Setting=zero-prior
2025.10
0
0
15
1.22
UITARS-1.5
Setting=zero-prior
2025.10
0
0
63
0.1
GPT-4o
Setting=zero-prior
2025.10
6.33
0
37
0.76
UITARS-1.5
Setting=prior-assisted
2025.10
11.33
1.33
92
0.12
GPT-4o
Setting=prior-assisted
2025.10
14.67
1
57
0.91
Claude3.7
Setting=prior-assisted
2025.10
19.33
3.33
92
1.18
BottomUp
Setting=zero-prior
2025.10
59.67
8.66
92
3.55
SAG-Agent
Setting=zero-prior
2025.10
107.33
13.33
93
3.03
Feedback
Search any
task
Search any
task