Researchers at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Lab (CSAIL) and Jameel Clinic, in collaboration with TechBio company Recursion, have released Boltz-2, a groundbreaking biomolecular foundation model. This next-generation AI tool, trained using Recursion's NVIDIA supercomputer, excels in modeling complex structures and predicting binding affinities with unparalleled speed and accuracy. Boltz-2 surpasses existing models like AlphaFold3 by combining structure and binding affinity predictions, achieving speeds up to 1000 times faster than traditional methods. Developed through a partnership between MIT and Recursion, Boltz-2 is now open-sourced under an MIT license, allowing for both academic and commercial use. The model's development was spearheaded by the Boltz team at MIT, with leadership from Professors Regina Barzilay and Tommi Jaakkola, offering scientists a powerful tool to advance drug discovery and explore new biological insights.