general

Normal Computing raises $50 million to cut AI chip cost and energy

Normal Computing secured $50 million to expand a chip design software business used by leading semiconductor firms and to advance a low-energy thermodynamic processor prototype.

Mar 25th 2026 · United States

Insights

  • Normal Computing raised $50 million in a round led by Samsung Catalyst with new investors including Galvanize, Brevan Howard Macro Venture Fund, and ArcTern Ventures.
  • The company’s software platform is already used by more than half of the top 10 semiconductor companies by revenue.
  • Normal has taped out a prototype chip that uses a thermodynamic approach, exploiting physical randomness to compute more efficiently than traditional GPUs.
  • The new funding will prioritize scaling the commercial chip design software while hardware work targets lower-energy inference for generative AI over the longer term.
  • Designing modern AI chips is increasingly expensive and risky, with some advanced designs costing more than $500 million before any units ship.