Volantis raises $88 million to expand AI chip memory capacity

  • Volantis raised $88 million to develop photonic links between AI chips and memory.
  • Its architecture targets 220 memory chips per GPU, versus about eight in traditional setups.
  • Volantis plans to deliver integrated A-1 inference engines to customers by 2027.

San Francisco semiconductor startup Volantis raised $88 million in a Series A on Oct. 1, 2026, bringing total funding to $97 million. Co-led by Lachy Groom and Abstract Ventures, the round included John Doerr and VXI Capital, following a $9 million seed round in 2025 backed by investors including Alex Wang and Trevor Blackwell. CEO Tapa Ghosh leads a team with experience at NVIDIA, AMD, Broadcom and Ayar Labs. Volantis is developing photonic interconnects based on custom vertical-cavity surface-emitting lasers, or VCSELs, to reduce memory-access constraints in AI inference. The company says its architecture could connect a GPU to as many as 220 memory chips, compared with about eight in traditional configurations. Its first system, A-1, is designed for models exceeding 20 trillion parameters, up to 10,000 tokens per second per user, about 10 terabytes of memory and 240 terabytes per second of bandwidth. Volantis plans to deliver integrated inference engines to customers by 2027.

The information on this website is generated using AI and we cannot guarantee its accuracy. Please use it as reference information only.