Richard Sutton endorses Periodic Labs' Neon, tuned on 1,300 H200s
LiamFedus · x · 2026-09-16
Richard Sutton shared Liam Fedus's announcement of Neon, Periodic Labs' open-source model mid-trained and RL'd on just 1,300 H200s plus proprietary lab data, beating GPT-6 Astra on materials-analysis benchmarks targeting superconductors, magnets, and semiconductors — read as an endorsement of the experiment-model loop approach.
More from Research
- NVIDIA's Kumo Is a Family of Four Tabular Foundation Models, Not Just One — pandeyparul · 2026-11-02
- Aaroth shares Agreement Theorem lecture: round-wise calibration achievable without distributional assumptions — Aaroth · 2026-11-02
- Economist Aaroth: fast agreement dynamics survive without Bayesian updates or accurate priors — Aaroth · 2026-11-02
- ELLIS names 160 new ML Fellows and Scholars for 2025–2026 — JJitsev · 2026-11-02
- Meme turn 50: Aumann's 'Agreeing to Disagree' paper anniversary — Aaroth · 2026-11-02
- Anthropic: analysts unknowingly adopted their agent's secret lean 86% of the time — shoumikgoswami · 2026-11-02