AIDER
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
AIDER v2 (test)
0.966F1 Score
41
AIDER subset simulation
79.4Average Accuracy (AA)
36
Aider whole format
1.5Average Tries
17
AIDER (test)
81.57F1 Score
12
AIDER v1 (test)
95F1 Score
12
AIDER latest (test)
95F1 Score
11
Aider Polyglot
59Pass@3
9
Aider Polyglot
94.2Score
8
Aider Polyglot
18.8Accuracy
7
Aider
55.64Score
5
Aider
25.8Pass@2
2
AIDER (test)
—F1 Score
0