AlphaFold

Generally Available
Google DeepMind
Google DeepMind
AlphaFold

Overview

DeepMind's protein-structure-prediction system. AlphaFold2 solved the decades-old protein-folding problem; AlphaFold 3 (co-developed with Isomorphic Labs) extends predictions to how proteins interact with DNA, RNA, ligands and other molecules. Its open database covers 200M+ structures.

The AlphaFold database has been used by 3M+ researchers across 190 countries and is the basis of the 2024 Nobel Prize in Chemistry. AlphaFold 3 is accessible via the AlphaFold Server for non-commercial research. In late July 2026 reports said DeepMind dismantled the dedicated AlphaFold team, reassigning most of the original paper's authors over the prior year to Gemini, enzyme design, fusion, genomics and Isomorphic Labs work as the lab pivots to a Gemini-centered, agent-based structure; several departing researchers, including 2024 Nobel laureate John Jumper, joined Anthropic.

Key Specs

Researchers

3M+ (190 countries)

Predicted Structures

200M+