Monday, 26 October 2026 Sign in

Microsoft Research open-sourced Agensh, a decentralized multi-agent harness for scaling coding work across up to 1,024 agents

Open source release Provisional 78% confidence first seen 25 October 2026, 03:32

Microsoft Research released Agensh, an open-source framework designed to coordinate large numbers of coding agents without a central orchestrator. Coverage describes scaling experiments on ProgramBench (up to 1,024 agents) and a reported increase in task pass rates as the number of agents grows.

Agensh ManyAgents Microsoft Research Anthropic Claude Code Codex GPT-6.1-Sol MarkTechPost ProgramBench

Source coverage

25 Oct, 03:32 Microsoft Open-Sources Agensh: A Multi-Agent Harness for 1,024 Agents MarkTechPost 25 Oct, 12:55 ManyAgents Product Hunt social

The daily briefing

Every AI story that matters,
in your inbox by 8am.

TLDRocket reads all relevant sources, removes duplicate coverage, and summarises the day in two minutes. Follow companies and topics for alerts, or get the briefing in Slack. Free, no spam, unsubscribe anytime.

TLDRocket 26 October 2026

The day’s biggest thread wasn’t another model release—it was money, and who gets it. Apollo Global Management’s latest analysis argues that, outside the biggest tech firms, AI has barely moved the needle: revenue per worker for the S&P 493 has hovered around $550,000 since 2019. Apollo’s implication is blunt: many companies are treating AI spend as a cost center until they can show clearer returns, which helps explain why mainstream adoption often looks like pilot purgatory rather than productive transformation.

Read the full briefing →

Microsoft Research open-sourced Agensh, a decentralized multi-agent harness for scaling coding work across up to 1,024 agents

Source coverage

Every AI story that matters, in your inbox by 8am.

Every AI story that matters,
in your inbox by 8am.