Instruction Following on AlpacaEval Arena-Hard (10% selection)
12.09AlpacaEval Score
InstructDiff
Updated 3mo ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.01 | 12.09 | 2.84 | 7.47 | |
2026.01 | 12.08 | 2.61 | 7.35 | |
2026.01 | 11.04 | 2.91 | 6.98 | |
2026.01 | 10.75 | 2.29 | 6.52 | |
2026.01 | 10.45 | 3.51 | 6.98 | |
2026.01 | 9.72 | 1.54 | 5.63 | |
2026.01 | 9.44 | 2.74 | 6.09 | |
2026.01 | 9.2 | 2.12 | 5.66 | |
2026.01 | 8.15 | 1.7 | 4.93 | |
2026.01 | 7.84 | 2.34 | 5.09 | |
2026.01 | 7.5 | 1.53 | 4.52 | |
2026.01 | 7.44 | 2.36 | 4.9 | |
2026.01 | 7.25 | 2.51 | 4.88 | |
2026.01 | 6.9 | 0.78 | 3.84 | |
2026.01 | 5.47 | 2.02 | 3.75 | |
2026.01 | 5.44 | 2.18 | 3.81 | |
2026.01 | 4.6 | 3.2 | 3.9 | |
2026.01 | 1.36 | 0.26 | 0.81 |