Ex-OpenAI exec's lab trains Neon on 1,300 H200s to beat GPT-6 Astra at materials analysis
LiamFedus · x · 2026-09-16
Liam Fedus's new company has built high-throughput materials labs in Menlo Park, closing the loop between experiments and models: labs generate fresh data, models learn from it, then suggest what to try next, targeting hard problems like superconductors, magnets, and semiconductor materials.
- Using only 1,300 H200s plus months of proprietary experimental data, they mid-trained and RL'd an open-source model into Neon, which surpasses GPT-6 Astra on their analysis benchmark
- The team argues the scaling direction is clear: bigger, smarter models with better simulations and higher-throughput labs will run increasingly autonomous physical experiments
- They released real lab footage and blog posts detailing the effort
It's a landmark demonstration of the model-plus-automated-lab approach to AI for Science.
More from Infra
- AI data center boom collides with cities scarred by big industry, now reaching Philadelphia — TechCrunch AI · 2026-09-16
- Small business seeks local model advice beyond Qwen on ZGX Nano AI station — Maschinhunt · 2026-09-16
- NVIDIA exec: global AI penetration only ~2%, Vera Rubin ramp still 'insanely hard' — BenBajarin · 2026-09-16
- Inference providers offload idle GPU risk by forcing throughput reservations — AAAzzam · 2026-09-16
- First M5 Ultra Geekbench leak: 52,516 multi-core, ~35% faster than M3 Ultra — jedisct1 · 2026-09-16
- Apple execs on podcast: 2nm silicon, doubled Neural Engine, edge AI future — BenBajarin · 2026-09-16