Ayar Labs joined NVIDIA's NVLink Fusion ecosystem on June 2, 2026. The silicon photonics startup will make its co-packaged optics compatible with NVIDIA's optical and SerDes technologies for hyperscale AI factories.
Key facts
- Ayar Labs joined NVLink Fusion ecosystem on June 2, 2026
- CPO solutions target bandwidth bottlenecks in AI factories
- Follows $500M Series E funding round
- Lightmatter also joined NVLink Fusion ecosystem
- CoreWeave brought up first Vera Rubin NVL72 system
Ayar Labs announced it has joined the NVIDIA NVLink Fusion ecosystem, making its co-packaged optics (CPO) optically and electrically compatible with NVIDIA's optical and SerDes technologies According to HPCwire. The move targets a fundamental scaling problem: as AI factories grow to larger GPU counts with heterogeneous architectures, data movement and power consumption become critical bottlenecks.
The Optical Imperative for Rack-Scale AI
Ayar Labs' CPO solution addresses scaling limits by bringing high-bandwidth, low-latency, power-efficient connectivity to NVLink Fusion architectures. This helps customers expand design headroom as bandwidth requirements grow and electrical constraints tighten, according to the company. By putting optics where it matters most, Ayar Labs' CPO enables flexible system architectures and efficient AI scaling beyond the limits of copper.
“AI infrastructure is being co-designed from the ground up, and customers need more options to scale performance efficiently as bandwidth continues to rise,” said Mark Wade, CEO at Ayar Labs. “By joining the NVIDIA NVLink Fusion ecosystem, we're introducing co-packaged optics as a foundational building block for customers deploying heterogeneous compute in NVIDIA AI factories.”
The collaboration is designed to help customers deploy heterogeneous compute, including custom silicon, within NVIDIA rack-scale platforms while preserving NVLink-based architecture investments. NVLink Fusion provides a path for integrating custom CPUs and XPUs into NVIDIA's rack-scale architecture and ecosystem, enabling customers to move faster from system concept to deployment at scale.
“NVLink Fusion, combined with Ayar Labs' CPO technology, gives customers more options to build heterogeneous AI factories,” said Ashish Karandikar, Vice President at NVIDIA. “By expanding the NVLink Fusion ecosystem with Ayar Labs' optical connectivity, innovators can scale bandwidth and bring heterogeneous AI infrastructure to market faster.”
The Competitive Landscape
Ayar Labs' announcement comes alongside Lightmatter joining the same NVLink Fusion ecosystem According to Business Wire. Both silicon photonics startups are racing to replace copper interconnects in AI clusters as bandwidth demands outpace electrical signaling. The timing is notable: CoreWeave just completed bring-up of the first NVIDIA Vera Rubin NVL72 system According to Data Center Dynamics, and Supermicro introduced DCBBS blueprints for the Vera Rubin NVL72 platform According to HPCwire.
Ayar Labs' CPO solutions complement NVIDIA's broader AI factory stack by pairing NVLink Fusion with NVIDIA's end-to-end networking platform. Together, these technologies support next-generation AI factories as they scale within and across racks, where bandwidth density, latency, and power efficiency directly impact utilization and total cost of ownership.
Today's announcement builds on Ayar Labs' successful close of its $500M Series E funding, which included participation from strategic investors. The company did not disclose the specific valuation or full investor list for the round.
What to watch
Watch for deployment timelines of Ayar Labs' CPO in Vera Rubin NVL72 systems from CoreWeave and other hyperscalers. The key metric is whether optical interconnects can deliver the promised bandwidth density improvements in production environments by Q4 2026.
Source: hpcwire.com
