Close-up of microchips on a circuit board — Photo by Jakub Pabis / Pexels

Volantis raises $88M Series A for photonic AI inference

Volantis lands $88M Series A to build A-1 photonic inference systems aiming at 20T+ parameter models and 10k tokens/sec per user.

THE HOOK

Another $88M bet that light, not copper, breaks the AI memory wall.

THE GIST

Volantis closed an $88M Series A co-led by Lachy Groom and Abstract Ventures (John Doerr, Dwarkesh Patel, Naveen Rao among participants). Its A-1 system targets models over 20 trillion parameters at up to 10,000 tokens/sec per user via photonic interconnects. First customer engines planned for 2027.

THE TAKE

Memory bandwidth is the real tax on giant models, and Volantis is selling an optical escape hatch. Micro-VCSELs and picojoule-per-bit links sound great on a press release. Shipping integrated inference engines in 2027 is the part that sorts hype from hardware.

OPINION

Founders from NVIDIA, AMD, Broadcom and Ayar Labs know the CoWoS and VCSEL supply chains — that’s the credibility. Still treat the PR as a fundraising milestone, not a shipping date. Watch who signs A-1 LOIs before the silicon story gets real.

  • Story type: Important story
  • Impact: Medium
  • Confidence: High (official announcement)
  • Audience: Business
  • What changed: Volantis raised $88M Series A for photonic AI inference targeting 20T+ models
  • What to watch: A-1 customer deliveries in 2027 and whether photonic memory bandwidth claims survive silicon
  • Reading time: 2 minutes