Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

AlpacaEval 2.0 on UltraFeedback

30LC

SSPO

2.5449.67216.823.928Oct 28, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.10
3020.7
2025.10
26.718.1
2025.10
20.720.8
2025.10
19.118.7
2025.10
18.816.4
2025.10
18.215.6
2025.10
18.112.9
2025.10
1813.6
2025.10
1712.8
2025.10
16.710.9
2025.10
16.718.2
2025.10
16.414.9
2025.10
16.213.3
2025.10
15.116.3
2025.10
1511.4
2025.10
14.913.2
2025.10
14.912.8
2025.10
14.816
2025.10
14.516.6
2025.10
14.415.8
2025.10
14.315.1
2025.10
13.415.3
2025.10
13.28.3
2025.10
1313.7
2025.10
1313.7
2025.10
12.112.6
2025.10
1010.9
2025.10
9.410.3
2025.10
7.74.3
2025.10
7.24.1
2025.10
4.93.1
2025.10
4.62.6
2025.10
4.42.6
2025.10
4.42.6
2025.10
4.32
2025.10
42.5
2025.10
42.5
2025.10
42.5
2025.10
42.6
2025.10
3.92.6
2025.10
3.72.3
2025.10
3.62.5