Can Open Models Keep Up With OpenAI and Anthropic? | Nathan Lambert

Explains why compute, data and RL environments, not only talent, set the open versus closed gap, which shapes which models engineers can rely on.

• distillation — Training a small, cheap model to imitate a big one's outputs, keeping much of the capability at a fraction of the cost.

• videoAvoiding vulnerabilities in AI code
• videoAI Is Learning to Hack. Faster Than We Expected.
• videoAgent Experience: Building an Open Web for the AI Era
• videoGoogle DeepMind Developers: How Nano Banana Was Made

• videoLocal Models: Trust, Control, Optimization — Carter Abdallah, NVIDIAAI Engineer
• articleOpen models recap: more on Kimi K3, Qwen 3.8, Xi's WAIC speech, distillation, the open-closed gap, and what's nextNathan Lambert
• videoWhy Specialized AI Could Beat The God Modela16z

Checking sign-in…

Loading comments…