130 Hours and $2.6M: The Real Cost of One RL Training Run

A viral post revealed that one RL training run took 130 hours, 75B tokens and about $2.6M, with tinygrad praising Xiaomi's reportedly impressive results and mocking US labs' cost efficiency, fueling debate over commoditized model training.

2026-09-23 ~ 2026-09-23 · 2 related posts

• One RL Run Cost 130 Hours, 75B Tokens and $2.6M — the Real Price of Scaling RL — burny_tech · 2026-09-23
• Xiaomi's $2.6M RL run: 130 hours, 75B tokens fuel 'commoditize training' debate — burny_tech · 2026-09-23