Liam Fedus team trains Neon with RL on lab data using 1,300 H200s, beats GPT-6 Astra
LiamFedus · x · 2026-09-16
Liam Fedus's team built high-throughput materials labs in Menlo Park creating a loop where experiments generate fresh data and models decide what to try next. Using only 1,300 H200s plus months of experimental data, they mid-trained and RL'd an open-source model into "Neon", surpassing GPT-6 Astra on their analysis benchmark. They also built pbox, an in-house RL sandbox for training trillion-parameter models, targeting superconductors, magnets, and semiconductor materials first.
Related event: Periodic Labs beats GPT-6 on materials science with just 1,300 H200s(41 posts)→
More from Infra
- Chris Lattner to keynote PyTorch Conference on Mojo and MAX for heterogeneous compute — PyTorch · 2026-09-16
- US data centers may use more natural gas than Germany and Japan combined by 2035 — TechCrunch AI · 2026-09-16
- Four US locales pause data-center builds in one week; Google, Amazon projects exempt — BenBajarin · 2026-09-16
- Researchers Gained Admin Access to Baseten's Production GitHub in 25 Minutes via PAT Takeover — bearsyankees · 2026-09-16
- Event-driven multi-agent architecture beat model routing on cost, 32-agent finance system shows — tradfyAi · 2026-09-16
- China narrows memory chip gap with Korea: 3 years in HBM, 1 year in NAND — robleclerc · 2026-09-16