One RL Run Cost 130 Hours, 75B Tokens and $2.6M — the Real Price of Scaling RL
burny_tech · x · 2026-09-23
zephyrz9 shared a striking RL training result: the team spent 130 hours of compute, 75B tokens, and roughly $2.6M on reinforcement learning to reach the reported result. The post underscores how expensive scaling RL has become — millions of dollars for a single training run.
Related event: $2.6M RL run nears SOTA, sparking training-cost debate(4 posts)→
More from Infra
- Omarchy 4.6 lets you share local AI models across your tailnet, says DHH — alexcovo_eth · 2026-11-12
- Eric Schmidt's foundation funds French NGO fighting against data centers — StephanSturges · 2026-11-12
- Epoch: Open-Weight GLM 5.3 Hacks Six Firefox Versions for ~$1,000 Each; Claude Fable 5.2 Sets ECI Record — Epoch AI · 2026-11-12
- Polymarket prices 28% odds of a state data center moratorium; Torus raises $500M — Polymarket · 2026-11-12
- AWS-wrapper cloud startup shuts down just one month after seed round — thedealdirector · 2026-11-12
- Utah startup Torus raises $500M for shipping container-sized AI data centers deployable in 4 months — Polymarket · 2026-11-12