Tencent Hunyuan Releases AI Research System Hyra 2.0
Tencent Hunyuan released the AI research system Hyra 2.0 on November 2, shifting its R&D focus 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. The system splits research work into three layers—Overall, Mission, and Worker—and designs three separate storage mechanisms: project state, layered Agent memory, and a shared knowledge base called Books. During project retrospectives, transferable methods can be organized into skills for use in subsequent tasks. Previously, Hyra 1.0 drove open-source models to solve an open problem in additive combinatorics that had remained unresolved for over fifty years.
Why read this
AI score 53
The original text explains Hyra 2.0's three-layer Agent architecture along with its memory and knowledge base design, allowing readers to understand how research Agents shift from solving problems to building tools.
Topics
Body · AI translation
Tencent Hunyuan Releases Hyra 2.0: AI Scientific Research Shifts from "Solving Problems" to "Building Tools"
Phoenix News Tech Download the App for Exclusive Early Access November 02, 2026 20:51:43 From Beijing
Phoenix News Tech (Editor/Song Shumei) On November 2, Tencent Hunyuan officially released its AI scientific research system Hyra 2.0, shifting its R&D focus from directly using AI to tackle scientific challenges to having general-purpose Agents participate in the R&D and upgrading of scientific models, research workflows, and software tools, driving a paradigm shift in scientific research from "doing practice problems" to "building tools."
Previously, Hyra 1.0 drove open-source models to solve an open problem in additive combinatorics that had been unresolved for more than fifty years, validating the feasibility of the approach of "giving AI difficult problems with clear evaluation criteria and having it try repeatedly." Building on this, Hyra 2.0 further proposes that the gains from tool upgrades can be repeatedly reused by downstream tasks, thereby forming an intelligence compound effect.
In terms of architecture, Hyra 2.0 splits research work into three layers: Overall, Mission, and Worker. Overall is responsible for communicating with human researchers and aligning goals; Mission is responsible long-term for a single research problem and judges trade-offs between approaches; and Worker independently reads code, runs experiments, analyzes the causes of failures, and can question Mission's judgments. The system also designs three separate storage mechanisms: project state, Agent hierarchical memory, and a shared knowledge base called Books. During project retrospectives, transferable methods can be organized into skills for direct use by subsequent tasks.