Third Moment Research scales 10,000 agents across 36+ hour runs
Source published ·Modelwire updated
Original coverage: OpenAI ↗·How Modelwire adds context
The development
Third Moment Research has demonstrated a significant scaling milestone by orchestrating 10,000 agents operating continuously for over 36 hours using OpenAI's latest models including GPT-6 Astra and GPT-6.1 Sol. This capability signals a shift toward persistent, long-horizon agent systems that improve through extended interaction rather than single-turn inference. The deployment across multiple specialized models (Codex for code, Daybreak for cybersecurity) suggests emerging patterns in how labs are partitioning complex tasks across purpose-built agents. For infrastructure and model providers, this validates demand for systems designed around agent persistence and multi-day coherence, reshaping expectations for production AI workloads beyond traditional chat and completion use cases.
Modelwire’s AI-generated summary of coverage from OpenAI.
Modelwire analysis
Skeptical readOur AI-generated reading of the wider context and the next developments to watch.
The announcement is published directly by OpenAI, meaning Third Moment Research's results have not been independently audited. Absent token consumption figures and error rates across that 36-hour window, the headline numbers describe scale without describing efficiency or reliability.
The efficiency question is not abstract. Research covered here on October 11 found that multi-agent systems consume five times more tokens for marginal quality gains, with performance plateauing beyond ten agents ('AI agent teams waste massive tokens for barely measurable quality gains'). Third Moment is running 10,000. That gap demands explanation, not just a deployment count. The Sophos-Daybreak case from October 9 is the stronger precedent for credible agent benchmarking: it named a specific workflow, a before-and-after time metric, and a human oversight layer. This announcement offers none of that structure. The Epoch AI and Anthropic findings from the same week also documented a persistent ceiling on agent self-validation, which matters for any claim about 36-hour autonomous operation.
If Third Moment publishes token consumption and task-completion accuracy data for this deployment within the next 60 days, the efficiency objection either collapses or gets confirmed. No follow-on disclosure would suggest this is closer to a reference architecture showcase than a production benchmark.
This interpretation is generated from the summary above and the archive coverage cited below. Our methodology · Report an error
Coverage behind this analysis
These archive entries ground the connection in our analysis. They are ordered by source publication date, with links to our coverage and the original sources.
·The Decoder
Multi-agent systems hit efficiency ceiling with minimal quality gains
Multi-agent AI systems are hitting a hard efficiency wall. New research from Vals AI shows that deploying teams of agents delivers marginal quality improvements over single agents while consuming five times more tokens. Anthropic's internal findings reinforce the pattern: beyond ten agents, performance plateaus entirely while computational costs continue climbing. This challenges a core assumption…
MentionsThird Moment Research · OpenAI · GPT-6 Astra · GPT-6.1 Sol · Codex · Daybreak
How this coverage is produced
Modelwire uses AI to generate summaries and context from source headlines, snippets, and selected archive coverage. Automated checks do not verify every claim, and items are not routinely reviewed by a person before publication. Zacaria Solis operates the site. Read the linked source for the full evidence and report errors through our corrections process.
Modelwire summarizes, we don’t republish. OpenAI originally reported this story as “How Third Moment Research runs 10,000 agents for 36+ hours”. The full content lives on openai.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.