deepseek-v4.1IMAGE

V4.1 Flash 1017 boosts wish-style coding for games and 3D from scratch

User reports that V4.1 Flash 1017, trained with self-feedback RL, improves building complete projects from scratch, especially game development and 3D modeling tasks.

岚叔

@LufzzLizArchived 10 days agoOriginally posted 2026.10.17

V4.1 Flash 1017 boosts wish-style coding for games and 3D from scratch (image 1 of 2)

V4.1 Flash 1017 boosts wish-style coding for games and 3D from scratch (image 2 of 2)

岚叔 @LufzzLizOriginally posted 2026.10.17

Open original on X

卧槽，果然周末来啊！这斩杀线无敌了哈哈，DeepSeek终于懂大众需求了：能够支持实现更多“许愿式”编程任务及项目开发请求： 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

This post quotes another post: x.com/lufzzliz/status/2111347869638946828

x.com/lufzzliz/status/2111378421460959373Archived 2026.10.18 16:49