Microsoft Research has released Agensh, an open-source framework for multi-agent systems that deviates from traditional centralized orchestrators. In this system, individual agents autonomously select their tasks and collaborate using shared resources like a Git repository, a chat channel, and a findings board. Benchmarking on ProgramBench demonstrated improved performance, with the mean test pass rate increasing from 19.31% with a single agent to 28.78% with 128 agents, indicating that larger agent teams achieve milestones more rapidly. AI
IMPACT This decentralized approach to multi-agent systems could offer new paradigms for complex task automation and coordination.
RANK_REASON Open-source release of a novel AI framework by a major research lab. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — mastodon.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →