Volantis announced an $88 million Series A on October 1, 2026, to develop and commercialize A-1, an AI inference system that the company says will use photonic links to connect compute chips with a larger pool of memory. Its headline specifications—support for models exceeding 20 trillion parameters and speeds of up to 10,000 tokens per second per user—are design targets, not independently verified results. Volantis plans to deliver its first integrated inference engines to customers in 2027.
What Volantis is building
Volantis describes A-1 as an inference system that uses a photonic fabric to link compute chips and memory. The company’s stated aim is to increase memory capacity and bandwidth together, rather than treating them as a fixed tradeoff. In a September 29, 2026, company post, founder Tapa Ghosh said: “We’re building a system for AI inference that uses photonics to break the memory wall.” That description reflects Volantis’ design rationale, not an independently established performance result.
As an Amazon Associate I earn from qualifying purchases.
In the company’s account, integrated micro-VCSELs are part of the optical fabric intended to pool memory while raising bandwidth. SiliconANGLE also reported specifications for the system and its optical links as claims attributed to Volantis; neither that coverage nor the company materials cited here establish independently measured, full-system A-1 performance.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →What the $88 million is for
The Series A was co-led by Lachy Groom and Abstract Ventures. Other participants named in the company announcement were John Doerr, VXI Capital, Triatomic, Susa Ventures, and angel investors Dwarkesh Patel, Naveen Rao, and Sholto Douglas. Volantis says the funding will support A-1 development and commercialization, expansion of its engineering team, and progress toward customer deployment. The company announced the round on October 1, 2026.
#1 Best Overall
How to read A-1’s targets
Volantis says A-1 is designed for models exceeding 20 trillion parameters and up to 10,000 tokens per second per user. These are company-stated design targets, not benchmark results that establish what a delivered system can achieve. The cited material does not provide an independent A-1 system benchmark or a separate published study validating either target.
To judge the system once it is delivered, readers will need measurement details as well as headline numbers: the model and workload used, what “per user” means under the test conditions, how many users or requests run concurrently, and whether reported speed reflects the complete inference system. Results should also clarify how memory capacity and bandwidth behave together in operation. Without a matched, independently measured basis, these targets cannot fairly be compared with demonstrated results from other inference hardware.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What happens next
Volantis says it plans to deliver its first integrated inference engines to customers in 2027. That is a forward-looking schedule, not evidence that delivery has occurred. Customer deployments and transparent system-level measurements will be important milestones for determining whether A-1’s photonic memory architecture achieves the capacity, bandwidth, and inference performance the company intends.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
