DeepSeek trained DeepSeek-V4.1-Flash-1017 using reinforcement learning with self-feedback to improve performance on end-to-end coding, game development, and 3D modeling tasks

From the articleDeepseek V4.1 Flash 1017 beats larger rivals in agentic tasks while once again massively cutting per-task costsAlso in DeepSeek Just Released V4.1-Flash-1017, Scores 77.0 on DeepSWE, Opens Search API