阶跃星辰发布 Step 5 Preview 技术博客，详解智能体任务的架构与训练设计

StepFun· @StepFun_ai · X·2026-10-15 21:00· 2 小时前AI 评分56

AI 导读

阶跃星辰发布技术博客《Inside Step 5 Preview: Architecture and Training for Agentic Work》，介绍架构与训练如何协同设计以支持更长、更复杂的智能体任务。

正文

Our technical blog, "Inside Step 5 Preview: Architecture and Training for Agentic Work," covers how architecture and training were co-designed for longer, more complex agent tasks:

• 92-layer narrow-and-deep architecture for implicit multi-hop reasoning
• Sparse GQA to cut million-token context cost, with 8-token block indexing reducing indexing and top-k overhead to 1/8
• FP32 precision on key states plus fine-grained gradient clipping for stable training
• Million-scale verifiable environments
• RL for long-horizon tasks with context compression built into policy training

https://platform.stepfun.ai/blog/inside-step-5-preview

来源：StepFun · x.com