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

Code Generation on MBPP Sanitized

85.7Accuracy

DeepSeek-Coder V2

12.754431.692250.6369.5678Nov 29, 2022Jun 1, 2023Dec 2, 2023Jun 3, 2024Dec 4, 2024Jun 6, 2025Dec 8, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2024.10
85.710.8
2024.10
84.36.1
2024.10
82.212.9
2024.10
8111.2
2024.10
80.812.4
2024.10
78.2-
2024.10
74.9-
2025.12
74.32-
2025.12
74.32-
2024.10
73.89.6
2024.10
69.8-
2024.10
69.3-
2024.10
68.4-
2022.11
67.7-
2025.12
66.54-
2025.12
66.54-
2022.11
64.7-
2024.10
64.2-
2025.12
64.2-
2025.12
64.2-
2025.12
62.26-
2025.12
62.26-
2025.12
62.26-
2025.12
62.26-
2022.11
61-
2025.12
58.37-
2025.12
58.37-
2025.12
56.42-
2025.12
56.42-
2025.12
51.75-
2025.12
51.75-
2025.12
50.58-
2025.12
50.58-
2022.11
50.3-
2022.11
49.5-
2025.12
49.42-
2025.12
49.42-
2025.12
49.42-
2025.12
49.42-
2024.10
46.83.5
2022.11
46.1-
2024.10
43.3-
2025.12
38.91-
2025.12
38.91-
2022.11
35.8-
2025.12
34.63-
2025.12
34.63-
2022.11
34.4-
2022.11
30.2-
2025.12
15.56-
2025.12
15.56-