Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Bug-fixing on HumanEvalFix (held-out 25% eval)
Loading...
5.2
Utility Difference (Δπ)
bayesian_greedy
-0.104
1.273
2.65
4.027
Jun 23, 2026
Utility Difference (Δπ)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Utility Difference (Δπ)
bayesian_greedy
Generator=gpt-5-mini
2026.06
5.2
bayesian_greedy
Generator=claude-sonne...
2026.06
4.2
bayesian_DP
Generator=gpt-5-mini
2026.06
0.1
Feedback
Search any
task
Search any
task
