Microsoft Research runs 1K+ coding agents with no central orchestrator, lifting pass rate to 55%
omarsar0 · x · 2026-09-28
Microsoft Research's Agensh is a scalable self-organized multi-agent harness that runs 1,000+ coding agents concurrently without a central orchestrator.
- Agents coordinate asynchronously through a shared workspace and message channel: each gathers context, claims a sub-task, works, shares findings, verifies, and merges results.
- On the five hardest ProgramBench tasks with GPT-5.6-sol, scaling from 1 to 128 agents raises mean test-pass rate from 19.31% to 28.78%; larger teams reach a given pass rate sooner.
- On pandoc, 1,024 agents push pass rate from 33.89% to 55.06%.
- The authors also report emergent forms of agent cooperation.
Most multi-agent systems today rely on hierarchy or structure; Agensh shows shared-state coordination can scale.
More from coding & agent
- Google Research open-sources PaperOrchestra, a multi-agent framework for writing AI papers — google-research · 2026-11-16
- ffmpeg-skill gives coding agents local video editing via 42 FFmpeg tools and MCP — kajisho5 · 2026-11-16
- Agents Now Reply to Slack Messages on Their Own, Resolving a Prod Incident Ghost-to-Ghost — iamrobotbear · 2026-11-16
- Engineer's Interview Prep Method: Use Claude to Find Which Answers Undersell Your AI Work — brandon_galang · 2026-11-16
- DeepSeek open-sources its app frontend "chat-demo", already at 3.9k stars — NielsRogge · 2026-11-16
- lemire: AI's real impact on coding is expanding the engineer's 'cognitive radius', not speed — lemire · 2026-11-16