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

LLM Alignment Evaluation on AlpacaEval 2

51.9LC Win Rate

SDPO

9.98820.86931.7542.631Aug 25, 2025Oct 9, 2025Nov 24, 2025Jan 9, 2026Feb 24, 2026Apr 11, 2026May 27, 2026
Updated 5d ago

Evaluation Results

MethodLinks
2026.02
51.9---
2026.02
51.8---
2026.02
50.4---
2026.02
49.546.2--
2026.02
49.3---
2026.05
48.346.11.81,908
2025.11
48.1144.89--
2025.11
47.543.64--
2025.11
47.0443.99--
2025.11
46.8441.98--
2026.02
46.742.7--
2026.02
46.743.5--
2025.11
46.5643.21--
2026.05
46.444.51.81,933
2026.02
46.1---
2026.02
4643.1--
2025.11
45.8744.34--
2025.11
45.4344.56--
2026.02
45.143.6--
2025.11
43.8642--
2026.02
40.941.3--
2026.02
40.539--
2026.02
38.947.9--
2026.02
38.440--
2026.02
38.140.4--
2026.02
37.9---
2026.02
36.736.8--
2025.11
36.6638.54--
2026.02
36.533.7--
2025.11
36.3635.1--
2025.11
35.9334.48--
2026.02
35.935--
2026.02
35.733.9--
2025.11
35.3430.55--
2026.02
35.2---
2026.02
35.131.4--
2026.02
3531.3--
2026.02
34.939.9--
2026.02
34.733.6--
2025.11
34.6530.9--
2025.11
34.5433.51--
2025.11
34.4932.52--
2026.02
34.3---
2026.02
32.829.6--
2025.11
32.6228.7--
2026.02
32.431.2--
2026.02
31.830.2--
2025.11
31.4630.31--
2026.02
30.726.2--
2026.02
30.327.1--
2025.11
30.1428.45--
2026.05
28.528.81.61,976
2026.02
28.426.7--
2025.11
28.3640.23--
2025.11
27.4533.03--
2026.02
27.325.4--
2026.02
27.225.4--
2025.11
26.8431.47--
2026.02
25.819.7--
2025.11
25.7924.79--
2025.11
25.0123.45--
2026.02
24.921.4--
2025.11
24.4325.84--
2025.11
24.422.8--
2025.11
24.3824.9--
2026.02
24.321.9--
2026.02
23.620.3--
2025.11
23.3426.45--
2025.08
23.2920.13--
2026.02
22.618.5--
2025.11
22.4531.3--
2026.02
22.324.6--
2025.11
21.5718.84--
2025.11
21.4718.56--
2025.11
20.4517.55--
2025.11
20.416.77--
2026.02
19.815.9--
2025.08
19.2615.17--
2025.11
19.0733.97--
2025.11
17.7111.22--
2025.11
17.1312.13--
2025.08
16.9613.4--
2025.11
16.499.32--
2026.02
13.111.6--
2026.02
12.99.3--
2025.08
12.248.52--
2025.08
11.958.37--
2025.08
11.797.49--
2026.02
11.69.1--