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

Code Evaluation on SWE Verified Agentless

57.6pass@1

DeepSeek-R1 0528 671B

19.01629.03339.0549.067Dec 15, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
57.6
53.8
2025.12
48.9
2025.12
43.6
2025.12
43.1
2025.12
37.2
2025.12
27.4
2025.12
20.5