San Francisco semiconductor startup Volantis has raised $88 million to develop technology aimed at improving how artificial intelligence processors connect with memory, Reuters reported on 1 October 2026.
The funding round was led by Lachy Groom and Abstract Ventures, with participation from investors including John Doerr, VXI Capital, Triatomic and Susa Ventures.
Volantis is targeting a challenge at the heart of AI infrastructure: getting enough data to powerful processors quickly enough. On its website, the company describes an approach that uses optical connections inside an accelerator to link computing components with larger pools of memory.
By carrying information using light, these connections are designed to reach further than conventional electrical connections. That could give chip designers more flexibility in how much memory they place alongside a processor, helping overcome physical constraints that limit existing designs.
The company is positioning its technology for AI inference, the process through which trained models generate responses. Its stated ambitions include running very large models with shorter response times and reducing the cost of generating output.
Those benefits remain company claims. The central proposition is that improvements in memory capacity and data movement could help AI systems make better use of their computing power.
Chief executive and co-founder Tapa Ghosh told Reuters that Volantis aims to deliver a chip in 2027, with AI coding among the potential applications.
