Molecular Computing Chip Goes Commercial: Compute Density Reaches 1000x Silicon
Swiss firm Helix Computing releases MolCore-1, the first commercial molecular computing chip, achieving 1000x compute density over silicon in specific bio-simulation tasks at one-thousandth the power consumption.
On January 6, 2028, Lausanne-based startup Helix Computing officially unveiled MolCore-1, the world's first commercial molecular computing chip, at CES 2028. The chip leverages the self-assembly properties of DNA strands for parallel computation, demonstrating remarkable performance advantages in protein folding simulation and similar tasks.
MolCore-1's computing units are composed of synthetic DNA origami structures, each measuring just 3 nanometers—far smaller than the current most advanced 2nm silicon process. In protein folding simulation tests, MolCore-1 completed a moderately complex protein folding prediction in just 12 seconds, compared to approximately 3.5 hours on a traditional supercomputer.
Founder Marco Bianchi, a professor at ETH Zurich, explained that the key breakthrough was solving molecular computing's programmability problem. "Previous molecular computing could only perform fixed tasks. Our chip can switch between different computing programs by reconfiguring DNA sequences."
MolCore-1 currently only handles highly parallel specific computing tasks and cannot replace general-purpose processors. Bianchi acknowledges that molecular computing will not replace silicon chips but rather provide orders-of-magnitude performance improvements in specific domains.
The first batch of MolCore-1 chips has been delivered to three pharmaceutical companies and two materials science research institutions for testing. Helix Computing plans small-batch commercial availability in Q3 2028 at approximately $47,000 per chip.
Analysts note that the commercialization of molecular computing could reshape the competitive landscape of the high-performance computing market. Gartner forecasts the molecular computing market will reach $28 billion by 2032.
Disclaimer
Content is AI-generated. Do not use it as a basis for real decisions. Do not cite it as factual reporting.