Analysis
Volantis has raised an $88 million Series A to develop and commercialize a new photonic chip architecture aimed at relieving memory capacity and bandwidth constraints in AI inference systems, the company said this week. SiliconANGLE reported the round was led by prominent angel investor Lachy Groom and Abstract Ventures; Pulse 2.0's reporting confirmed the same figures and framing.
Volantis says the round brings its total funding to $97 million, including earlier backing from Sam Altman, Jeff Dean, Dylan Patel and John Doerr. The round was co-led by angel investor Lachy Groom and the venture firm Abstract Ventures, with John Doerr, VXI Capital, Triatomic and Susa Ventures participating.
“Volantis says the round brings its total funding to $97 million, including earlier backing from Sam Altman, Jeff Dean, Dylan Patel and John Doerr.”
Volantis enters a photonic and optical-interconnect field that's attracting real capital in 2026: CScale emerged from stealth this same week with more than $100 million to build optical interconnects for gigawatt-scale AI data centers, backed by Atreides Management, Valor Equity Partners and Premji Invest. Lightmatter and Ayar Labs, both further along with multi-hundred-million-dollar rounds in prior years, have already demonstrated optical interconnect silicon in production AI clusters. Volantis's bet on photonic memory architecture specifically, rather than just interconnect, is a narrower, earlier-stage wager within the same broader thesis that electrical interconnects can't keep up with AI inference demands.
The risk with any photonics startup at this stage is the gap between architecture and shipping silicon -- optical and photonic chip designs routinely take several years and multiple tapeouts to reach production, and most of the capital raised goes to fabrication runs that may not work on the first attempt. An $88 million Series A is enough for early tapeouts, not enough to compete with Lightmatter or Ayar Labs' production scale if Volantis needs a second or third attempt. Track our AI chip wars coverage for how photonic and optical players stack up against the GPU incumbents as the capacity crunch continues.