Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Countdown on Countdown (test)
Loading...
57.7
Accuracy
RLVR (with verifier)
-0.228
14.811
29.85
44.889
Nov 26, 2025
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
RLVR (with verifier)
Reasoning budget=2048...
2025.11
57.7
RARO
Reasoning budget=2048...
2025.11
54.4
SFT
Reasoning budget=2048...
2025.11
40.7
Iterative DPO
Reasoning budget=2048...
2025.11
40.4
Rationalization
Reasoning budget=2048...
2025.11
12.5
RL-Logit
Reasoning budget=2048...
2025.11
2.2
Base
Reasoning budget=2048...
2025.11
2
Feedback
Search any
task
Search any
task