Volantis raises $88M to to build photonic memory layer for next-gen AI (techfundingnews.com)

Volantis, a San Francisco-based semiconductor startup, has secured $88 million in Series A funding to develop its innovative photonic architecture, the A-1 system, aimed at enhancing AI inference capabilities. Co-led by Lachy Groom and Abstract Ventures, the funding will support the company’s mission to create a platform capable of running models that exceed 20 trillion parameters at speeds of up to 10,000 tokens per second per user by 2027. This initiative comes in response to the growing demands AI workloads place on existing memory infrastructures. The significance of Volantis’ work lies in its potential to overcome the limitations of current computing architectures, which often require compromises between high bandwidth and ample memory capacity. By employing a photonic interconnect that aggregates bandwidth as memory is added, Volantis aims to provide a unified memory solution that scales without sacrificing performance. Utilizing custom micro-VCSELs, the company’s approach consumes less power while ensuring a more efficient data movement between compute chips and memory systems. As the race for advanced AI infrastructure intensifies, Volantis’ focus on memory solutions positions it uniquely within a highly competitive landscape, drawing interest likened to other notable companies in the photonic computing space.