Researcher Nathan Lambert has left the Allen Institute for AI after contributing to the Olmo model series. His departure follows a tenure focused on open-source LLM development and alignment. This move marks a personnel shift within one of the few remaining non-profit research labs producing high-performance base models for the community.