https://derekwang2002.github.io/ai-intelligence/en/projects/third-moment-agent-system/ Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations. en Mon, 02 Nov 2026 16:56:00 GMT https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261014-01/ ev-20261014-01-published Wed, 14 Oct 2026 15:57:53 GMT New WATCH <p>An OpenAI customer story says Third Moment Research runs up to 10,000 concurrent agents for 36 to 48 hours, consuming roughly one trillion OpenAI tokens per day. Its default worker is Codex inside Docker, while outer GPT-6 Sol and GPT-6.1 Sol agents retry, terminate or escalate stalled work. The article is a vendor-hosted customer account and provides no raw telemetry or independent validation.</p><p><strong>Why read: </strong>Third Moment&apos;s 10,000-agent long-running account offers a rare supervision-architecture signal, but missing raw telemetry keeps it at WATCH.</p><p>Recommendation: WATCH</p><p>Sources: <a href="https://openai.com/index/third-moment-research/">How Third Moment Research runs 10,000 agents</a></p>