Microsoft Research open-sources Agensh, a self-organizing harness that scales to 1,024 coding agents

Core Background & Industry Pain Points

Most multi-agent coding harnesses rely on a central orchestrator to assign work, and that orchestrator caps how far a team can scale. Microsoft Research posted the Agensh preprint (arXiv 2609.26781) on Sep 22 and open-sourced the code on Oct 24. This is an update of our September story: some mechanisms and figures in the earlier version could not be traced to the paper or repo and have been rewritten from official sources.

Architecture Highlights & Internals

Agensh has no central orchestrator. Every worker runs the same loop: gather context, claim and self-assign a sub-task, act, verify, and merge progress asynchronously until the goal is met. Three pieces of infrastructure support it: Gitea as the shared workspace for proposed, ongoing and completed work; Mattermost as the message interface; and shared context for reusable findings and work intentions, adapted from DeLM. The paper reports that peer coordination, integration workflows and specialized roles emerge as the team grows.

Authoritative Benchmarks & Measured Scores

All figures are author-reported with no third-party reproduction yet. On the 5 hardest ProgramBench tasks (rebuild reference programs from behavior and bundled docs; GPT-5.6-sol high, 6h, no Internet), the mean final test-pass rate rises from 19.31% with 1 agent to 28.78% with 128 agents. On pandoc it climbs from 33.89% to 50.94% (128 agents) and 55.06% (1,024 agents). Larger teams also reach comparable pass rates sooner.

Developer Hands-on Guide

The repo includes a reproduction guide, a team launcher and a single-agent launcher; a Docker Compose single-node setup lets you compare a team against the single-agent baseline. Experiments depend on Copilot ACP sessions and GPT-5.6-sol, and compute or API cost at large scale is not disclosed, so start with small team sizes.