88Z AI, 28 July — Renmin University of China and Ant Group have jointly launched SearchOS, an open-source multi-agent search framework designed to provide system-level collaboration infrastructure for long-horizon complex information retrieval. By abstracting task progress, evidence relationships, and completion status into a system state shared across agents, the framework enables coordinated division of labor under unified scheduling, effectively preventing redundancy and memory loss.
At a glance
- Renmin University and Ant Group open-source SearchOS framework for long-horizon complex retrieval collaboration
- Unified scheduling via shared system state prevents redundancy and memory loss among multi-agents
The update
Renmin University of China and Ant Group have collaborated to launch SearchOS, an open-source multi-agent search framework. This innovative tool is designed to address collaboration challenges in long-horizon complex information retrieval by providing system-level infrastructure support. The core mechanism of SearchOS lies in abstracting key elements of the search process, including task progress, logical relationships between evidence, and information completion status, into a system state shared by all agents. This design allows multiple agents to perform efficient division of labor and collaboration under unified scheduling. By sharing states, the system significantly reduces redundant work and overcomes the memory loss issues common in traditional multi-agent systems, thereby improving overall retrieval efficiency and accuracy.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- 机器之心Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.