Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Preference Elicitation on SpecBench Task 1
Loading...
74
Improvement Rate (T0->T5)
Buddy
49.04
55.52
62
68.48
May 13, 2026
Improvement Rate (T0->T5)
Improvement Rate (T5->T10)
Improvement Rate (T0->T10)
Query Count (T5 Budget)
Query Count (T10 Budget)
Early Submission Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Improvement Rate (T0->T5)
Improvement Rate (T5->T10)
Improvement Rate (T0->T10)
Query Count (T5 Budget)
Query Count (T10 Budget)
Early Submission Rate
Buddy
2026.05
74
77
92
-
-
0
Claude Code
Backbone=Opus-4.6
2026.05
62
67
80
-
-
1
Cursor CLI
Backbone=GPT-5.4-mini
2026.05
62
61
78
-
-
3
Gemini CLI
Backbone=Gemini-2.5-pro
2026.05
50
57
61
-
-
17
Feedback
Search any
task
Search any
task