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

Code Generation on HumanEval (Speedup vs. AR)

1.91Speedup vs AR

HiSpec

1.01561.24781.481.7122Oct 1, 2025
Updated 7d ago

Evaluation Results

MethodLinks
2025.10
1.91
2025.10
1.84
2025.10
1.71
2025.10
1.69
2025.10
1.67
2025.10
1.67
2025.10
1.54
2025.10
1.52
2025.10
1.51
2025.10
1.5
2025.10
1.36
2025.10
1.32
2025.10
1.24
2025.10
1.24
2025.10
1.23
2025.10
1.19
2025.10
1.18
2025.10
1.12
2025.10
1.05