Tencent Hunyuan releases AI research system Hyra 2.0, shifting toward having Agents develop reusable tools
English text is being prepared.
Tencent Hunyuan releases AI research system Hyra 2.0, with the R&D focus shifting from directly using AI to tackle scientific problems toward having general-purpose Agents participate in the development and upgrading of scientific models, research workflows, and software tools, so that the results can be reused repeatedly in subsequent research. Architecturally, Hyra 2.0 splits research work into three layers: Overall, Mission, and Worker; Workers can independently read code, run experiments, analyze the causes of failures, and question Mission's judgments. The system also designs three separate storage systems: project state, Agent layered memory, and a shared knowledge base called Books. During retrospectives, transferable methods can be organized into skills for use in subsequent tasks.
Why read this
AI score 60
The original text explains Hyra 2.0's shift from solving problems to building tools, along with its three-layer Agent architecture, which can be used to understand the division-of-labor design of research agents.
Topics
Body
腾讯混元发布 Hyra 2.0：为模型研发打造可复用工具
2026-11-02 21:08 IT之家 - 远洋
IT之家 11 月 2 日消息，今日腾讯混元发布了 AI 科研系统 Hyra 2.0，将研发重点从直接用 AI 攻克科学难题，转向让通用 Agent 参与科学模型、研究工作流及软件工具的研发与升级，让一次研发的成果在后续研究中被反复使用，推动科研范式从“刷题解题”走向“造工具”。
此前，Hyra 1.0 曾驱动开源模型解决加法组合学中悬置五十余年的开放问题，验证了“给 AI 明确评判标准的难题并循环尝试”这一路径的可行性。Hyra 2.0 在此基础上进一步提出，工具升级带来的收益可被下游任务反复复用，从而形成智能复利。
IT之家注意到，架构上，Hyra 2.0 将研究工作拆分为 Overall、Mission 和 Worker 三层，Overall 负责人与人类研究者沟通并对齐目标，Mission 长期负责单一研究问题并判断路线取舍，Worker 则独立读代码、做实验、分析失败原因，并可质疑 Mission 的判断。系统还设计了项目状态、Agent 分层记忆与共享知识库 Books 三套分离的存储机制，项目复盘时可将可迁移方法整理为 skill，供后续任务直接调用。