deepseek-v4.1IMAGE
DeepSeek V4.1 Flash 1017 uses self-feedback RL for from-scratch project builds
The author argues self-feedback reinforcement learning lets V4.1 Flash 1017 build complete projects from scratch in game dev and 3D modeling, enabling wish-style coding.
Max For AI
@MaxForAIArchived 10 days agoOriginally posted 2026.10.17
Max For AI @MaxForAIOriginally posted 2026.10.17
看起来@deepseek_ai 成为了中国模型团队里第一个部分跑通RSI的团队！ DeepSeek V4.1 Flash 1017 通过基于自反馈的强化学习（RL）训练技术，提升了在游戏开发、3D 建模等方面从零开始构建完整项目的能力，能够支持实现更多“许愿式”编程任务及项目开发请求。 让我们加速进化吧🚀
This post quotes another post: x.com/deepseek_ai/status/2111375640717033777
x.com/maxforai/status/2111381500063306098Archived 2026.10.18 16:48