OpenAI burned millions in compute on $1M prize: 10,000 agents, 88 hours, 130B tokens
Hesamation · x · 2026-09-09
Hesamation reveals OpenAI spent millions of dollars in compute to solve a problem with a $1M prize:
- They used an internal model "significantly more capable than GPT-6 Astra"
- Spawned 10,000 concurrent agents
- Ran for 88 hours
- Generated 130 billion output tokens
Even juicier, OpenAI responded: they saw none of the users' work, "but we CANNOT RULE OUT the possibility that their chats improved our model" — prompting the quip "own your weights."
Related event: OpenAI's Costly Navier-Stokes Run Sparks Scooping and Brute-Force Debate(26 posts)→
More from Models
- Mathematicians debate AI math writing: abysmal today, improving shockingly fast — miniapeur · 2026-10-27
- OpenAI launches GPT-6.1 Terra focused on computer use, at a fraction of the cost — minchoi · 2026-10-27
- The trust paradox: only the least trustworthy LLMs will do the sketchiest things — AaronBergman18 · 2026-10-27
- Claude's refusal to type one-time passwords is a real security impediment — AaronBergman18 · 2026-10-27
- repligate argues context full of noise is genuinely 'hard' for LLMs, mulls an info-theory model of breaks — repligate · 2026-10-27
- Fork builds engine to run Qwen at ~400 tok/s on three used V100s — jfinkpottery · 2026-10-27