Microsoft Research unveils Quine, a biology multimodal model

On September 29, 2026, Microsoft Research introduced Quine, describing it as a multimodal world model of biology. The project is presented as an early-stage research effort rather than a finished product, and its stated aim is to connect insights across biological scales and modalities so that biological knowledge is represented as one connected whole. Microsoft Research frames the motivation directly: biology does not operate in silos, and neither should the AI representation of it.

According to the announcement, Quine is meant to help scientists computationally search a hypothesis space far larger than intuition alone can cover, and to prioritize hypotheses before they reach the lab. The published excerpt also refers to experimental results, but it is cut off at that point. No specific results, benchmarks, datasets, or comparison figures were provided, and the announcement likewise did not mention model sizes, training data, or any way for outside researchers to access the system.

As of this announcement, Quine remains an early-stage research effort. Microsoft Research did not give a release timeline, a productization plan, or any details about external availability, so the project currently stands at the point of initial disclosure with no follow-up milestones reported.