# Agensh coordinates up to 1,024 coding agents without a central orchestrator

A Microsoft Research paper describes Agensh, a self-organized harness that coordinates parallel coding agents through shared state. With GPT-5.6-sol, scaling from 1 to 128 agents raises the mean final test-pass rate on the five hardest ProgramBench tasks from 19.31% to 28.78%; on pandoc, 1,024 agents raise it from 33.89% to 55.06%.

Language: en
Time zone: America/Los_Angeles
Current time (UTC, cached as of): 2026-11-17T15:13:00.000Z
Query date (America/Los_Angeles): 2026-11-17

HTML: https://didcodexreset.com/news/5e399770634f47b1951a6509.html

Content language: en
Localization state: sameLanguage

Source: elvis · Published 9/27/2026, 18:13:58

Microsoft Research’s Agensh is a self-organized multi-agent harness with no central orchestrator. Parallel coding agents coordinate through shared state: each gathers context, claims a sub-task, does the work, shares what it found, verifies the result, and merges it asynchronously through a shared workspace and a message channel. The setup was tested with more than 1,000 coding agents running at once.

On the five hardest ProgramBench tasks with GPT-5.6-sol, increasing the agent count from 1 to 128 raises the mean final test-pass rate from 19.31% to 28.78%. Larger teams also reach a given pass rate sooner. On pandoc, 1,024 agents take the test-pass rate from 33.89% to 55.06%. The authors also report forms of cooperation that the agents start on their own and that become standard practice as the team grows.

The paper is available on [arXiv](https://arxiv.org/abs/2609.26781).

Tags: Agensh, Microsoft Research, Multi-agent, ProgramBench, GPT-5.6-sol

[View original post](https://x.com/omarsar0/status/2104377054829613473)
