PulseAugur
EN
LIVE 06:58:18

Microsoft Research open-sources Agensh for 1,024 AI agents

By PulseAugur Editorial · [1 sources] ·

Microsoft Research has open-sourced Agensh, a new multi-agent framework designed to coordinate large numbers of AI agents without a central orchestrator. This decentralized approach allows individual agents to pick their own sub-tasks, share findings, and merge code into a common repository. Experiments showed that scaling Agensh from 1 to 1,024 agents significantly improved performance on complex coding tasks, with scores increasing and tasks being completed more rapidly as the number of agents grew. AI

IMPACT Enables more scalable and efficient coordination of large AI agent teams for complex tasks.

RANK_REASON Open-source release of a framework for coordinating multiple AI agents.

Read on MarkTechPost →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Microsoft Research open-sources Agensh for 1,024 AI agents

How we ranked this

Signal score
20 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
Open-source release of a framework for coordinating multiple AI agents.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. MarkTechPost TIER_1 English(EN) · Asif Razzaq ·

    Microsoft Open-Sources Agensh: A Multi-Agent Harness for 1,024 Agents

    <p>Microsoft Research open-sourced Agensh, a multi-agent harness with no central orchestrator. Self-organizing workers share a Git repo, chat and a findings board, scaling from 1 to 1,024 agents on ProgramBench under a 6-hour budget.</p> <p>The post <a href="https://www.marktechp…

RELATED ENTITIES

RELATED TOPICS