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

Multi-turn Dialogue Generation on MT-bench

5.62Mean Accepted Length

BlockPilot

1.6696642.6952323.72084.746368May 11, 2026May 19, 2026May 27, 2026Jun 5, 2026Jun 13, 2026Jun 21, 2026Jun 30, 2026
Updated 24d ago

Evaluation Results

MethodLinks
2026.06
5.622.85
2026.06
5.593.31
2026.06
5.292.69
2026.06
5.233.1
2026.05
4.52294.52
2026.06
4.372.55
2026.06
4.252.48
2026.06
4.232.48
2026.06
4.062.37
2026.06
3.911.99
2026.06
3.792.24
2026.06
3.792.23
2026.05
3.75973.89
2026.06
3.752.2
2026.05
3.72712.82
2026.05
3.69383.69
2026.06
3.452.01
2026.06
3.281.93
2026.06
3.221.91
2026.06
3.191.62
2026.05
3.15352.37
2026.05
3.07532.33
2026.05
3.07122.35
2026.05
3.0523.16
2026.06
3.021.74
2026.06
2.991.77
2026.06
2.851.45
2026.06
2.831.63
2026.06
2.741.62
2026.06
2.71.58
2026.05
2.68022.69
2026.06
2.581.51
2026.05
2.51511.9
2026.05
2.51152.52
2026.05
2.48631.92
2026.06
2.451.44
2026.05
2.29282.3
2026.05
1.82391.86
2026.05
1.82161.86