MIT Lincoln Laboratory has deployed an AI supercomputer at its Supercomputing Center in Holyoke, Massachusetts.
According to MIT, the system, dubbed TX-GAIN, is the most powerful AI-dedicated supercomputer deployed by a US university.
Powered by more than 600 Nvidia H100 GPUs, the system ranked 114 on the most recent edition of the Top500 list, achieving a HPL score of 13.39 petaflops, although the university claims it also has 2 exaflops of AI compute power.
MIT will use the supercomputer to support biodefense, materials discovery, and national security research. The data center where TX-GAIN is housed also supports collaborations with other MIT institutions, including the Haystack Observatory, the Center for Quantum Engineering, and the Air Force–MIT AI Accelerator.
“TX-GAIN will enable our researchers to achieve scientific and engineering breakthroughs,” said Jeremy Kepner, Lincoln Laboratory Fellow and head of the Supercomputing Center. “The system will play a large role in supporting generative AI, physical simulation, and data analysis across all research areas.”
Rafael Jaimes, a researcher in the Counter–Weapons of Mass Destruction Systems Group, added: “TX-GAIN is allowing us to model not only significantly more protein interactions than ever before, but also much larger proteins with more atoms. This new computational capability is a game-changer for protein characterization efforts in biological defense.”