OpenAI's Navier-Stokes run used a model just 4 days into training, still unfinished
haider1 · x · 2026-09-09
A wild detail about OpenAI's Navier–Stokes result: the internal model had only been training for 4 days when the 88-hour solving run began. A day or two later the team switched to an even better checkpoint that helped them "solve" it — and the model still hasn't finished its initial training run, implying its reasoning capabilities are still climbing fast.
Related event: OpenAI Claims 10,000 Agents Solved Navier-Stokes in 88 Hours(13 posts)→
More from Research
- HarnessMM cuts agent tokens per task by up to 88% while raising success rates — omarsar0 · 2026-10-26
- Ben Recht: The world needs a truly open language model, and it's easier to build than you think — beenwrekt · 2026-10-26
- New Blog Breaks Black-Box Distillation Into Five Practical Steps — chrisrohlf · 2026-10-26
- DetectorScope turns black-box AI text detectors into auditable ones with sparse autoencoders — _akhaliq · 2026-10-26
- OPST paper shows LLMs self-improve with zero human data, guided only by topic descriptions — _akhaliq · 2026-10-26
- SJTU's VaM lets robots learn online from mistakes, 95%+ success over 4-hour autonomous runs — chris_j_paxton · 2026-10-26