Periodic Labs trains Neon on just 1,300 H200s to beat GPT-6 Astra on materials analysis
LiamFedus · x · 2026-09-16
Liam Fedus's Periodic Labs built high-throughput materials labs in Menlo Park that close the loop between experiments and models: the labs generate fresh data, the models learn from it and guide what to try next. Using only 1,300 H200s plus months of proprietary experimental data, the team mid-trained and RL'd an open-weight model (Kimi K2.6) into Neon, a 1T-parameter model that surpasses GPT-6 Astra on their FrontierXRD analysis benchmark. First targets include superconductors, magnets, and semiconductor materials.
Related event: Periodic Labs beats GPT-6 on materials science with just 1,300 H200s(41 posts)→
More from Research
- AgentIR embeds agent reasoning traces into retrieval, plus a BrowseComp-Plus benchmark — CShorten30 · 2026-09-16
- AI Evals FAQ grows to 48 Q&As: sensitive data, huge traces, and stale gold datasets — HamelHusain · 2026-09-16
- TurnTrout Offers Shard Theory Explanation for Assistant Behaviors Shaped by AI-Free Documents — dhadfieldmenell · 2026-09-16
- NVIDIA releases FoundationPose on Hugging Face: unified 6-DoF pose model, no fine-tuning needed — _akhaliq · 2026-09-16
- 3 weeks through Stanford CS329A: the generator has outrun the verifier — le_james94 · 2026-09-16
- DeepSeekMath-V2 makes verification the product, scaling verifier compute ahead of the generator — le_james94 · 2026-09-16