Google's AI DeepMind, has introduced an advanced version of its AI model, AlphaFold 3, to predict life molecules, such as DNA, RNA, and smaller molecules known as "ligands" other than protein structures.

The model leverages a diffusion technique to generate a complex 3D model of molecular structure, as given by researchers. It is limited to academic and non-commercial users in the medicine, agriculture, materials science, and drug discovery industries.

Furthermore, the model and the research platform AlphaFold Server will be available to some researchers for free.

The company claims that the AlphaFold Server, powered by AlphaFold 3, can be used with limited computation power access and improves prediction accuracy by 50% compared to previous versions. 

Analyst QuickTake: This news comes just a week after Google and DeepMind introduced Med-Gemini , an AI model focusing on medicine to enhance the understanding and processing of medical data. It has also developed other medical models like MedLM that cater to healthcare needs. These developments could potentially assist medical researchers and clinicians by improving efficiency, enhancing communication, streamlining workflows, supporting research endeavors, and offering flexibility in addressing specific healthcare challenges.