Third Moment Research describes running 10,000 agents for 36+ hours
Overview
Third Moment Research says it runs 10,000 agents for more than 36 hours, according to a first-party OpenAI News post dated October 14, 2026.
The post names GPT-6 Astra and GPT-6.1 Sol as models the lab uses, alongside Codex and Daybreak for cybersecurity work. The source describes the lab's agents as built to focus for days and improve with use. The post is OpenAI's account of Third Moment's work, and no independent source has been reported.
Written by AI from the articles below · updated Oct 14, 12:06 PM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- OpenAI NewsOfficialHow Third Moment Research runs 10,000 agents for 36+ hours
AIThird Moment Research says it runs 10,000 agents for more than 36 hours, using GPT-6 Astra and GPT-6.1 Sol alongside Codex and Daybreak for cybersecurity. The source describes the lab's agents as built to focus for days and improve with use.
Heat trend
Not enough continuous observations to show a trend yet.