AMD, NVIDIA, OpenAI & Others Form An Optical Scale-up Consortium

Written by Michael Larabel in AI on 12 March 2026 at 09:19 AM EDT. 7 Comments
AI
AMD, Broadcom, Meta, Microsoft, NVIDIA and OpenAI jointly announced today the formation of the Optical Compute Interconnect (OCI) Multi-Source Agreement (MSA) optical scale-up consortium.

This consortium is focused on a multi-vendor, open ecosystem for optical scale-up interconnects. AI, of course, being the big focus of the optical interconnects in 2026.
"As large language models (LLMs) advance toward super intelligence, traditional copper-based connectivity is reaching limitations in physical reach which are impacting AI cluster scale-up domain architectures. OCI will enable migration from copper-based to optical-based scale-up architectures, alleviating copper interconnect bottlenecks.

The OCI specification, which is available by request at www.oci-msa.org, is architected to be power, latency and cost optimized. It combines non-return to zero (NRZ) modulation and wavelength division multiplexing (WDM) optical technology and shifts the connectivity paradigm from a module-centric to a silicon-centric model. By enabling tighter integration of optics with compute and networking silicon, OCI unlocks meaningful gains in bandwidth density and system scalability while meeting the aggressive power targets of legacy copper-based connectivity."

The Optical Compute Interconnect (OCI) Multi-Source Agreement (MSA) technical specification can be downloaded at OCI-MSA.org.

OCI MSA logo


It's interesting to see NVIDIA involved in this consortium with other vendors while notably absent is Intel.
"The OCI MSA provides a scalable open specification roadmap for the full AI rack supply chain, enabling multi-vendor optical PHY and interconnect deployments across several generations of hardware:

Standardized High-Density Interfaces: Promoting the OCI GEN1 4λ x 50Gbps NRZ (200Gbps/direction) and the OCI GEN2 400Gbps/direction bidirectional (BiDi) technology, achieving up to 800Gbps per fiber.

Massive Scalability: A roadmap scaling wavelength counts and data rates to reach 3.2Tbps per fiber and beyond. Enabling scale-up domains to be realized with both higher GPU counts and higher bandwidth per GPU.

Interoperable Form Factors: Support for pluggable, on-board, and co-packaged optics (CPO).

Efficiency at Scale: Enabling optical solutions to meet the aggressive performance, power and cost targets previously only associated with copper connectivity while providing significantly greater reach."

Those wanting to learn more can do so via today's press release.
Related News
About The Author

Michael Larabel is the principal author of Phoronix.com and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 20,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and OpenBenchmarking.org automated benchmarking software. He can be followed via Twitter, LinkedIn, or contacted via MichaelLarabel.com.

Popular News This Week