MemryX and Lenovo Advance Sovereign Edge AI in Saudi Arabia

Riyadh, Saudi Arabia — September 15, 2026

MemryX has signed a memorandum of understanding with Lenovo to expand the deployment of sovereign edge AI solutions throughout Saudi Arabia, building on joint platforms already running in the Kingdom.

The agreement creates a framework for the two companies to pursue opportunities together across infrastructure, industrial operations, smart cities and video management — segments where video and sensor data is generated continuously, and where moving all of it to a centralised cloud is neither practical nor desirable.

Rather than starting from a pilot, the MOU formalises work that is already in production. Two customer deployments pairing MemryX accelerators with Lenovo edge servers are operating in Saudi Arabia today, and the framework is designed to take that pattern to more customers, more sites and larger rack-scale configurations.

The joint platform

The combined solution pairs the Lenovo ThinkEdge SE455 V3 rugged edge server with the MemryX Cascade 100P AI accelerator, processing video and sensor data on-site instead of shipping it upstream. The result is lower latency on every inference, tighter control over where data lives, and far less dependence on centralised, power-hungry AI infrastructure.

“Together, we are delivering high-density, power-efficient AI inference close to customer operations” — and establishing a scalable platform for what comes next.

Already live in the Kingdom

Two deployments are in operation today, both running on existing camera infrastructure rather than requiring customers to rip and replace what they already own.

“What matters to customers is being able to do that while keeping their data close to where it’s created and under their control.”

What the MOU covers

The memorandum sets out four lines of joint work, moving from the deployments running today toward larger configurations:

• Expanding customer deployments — taking the ThinkEdge plus Cascade pattern to additional sites and sectors across the Kingdom.
• Validation through a Lenovo Center of Excellence — proving the joint platform under Lenovo’s own test and qualification programme.
• Solution development via Lenovo AI Innovation — building additional vertical solutions on top of the validated stack.
• Rack-scale server configurations — exploring denser deployments for computer vision inference, where many accelerator cards serve a large camera estate from a single rack.

Aligned with Vision 2030

Saudi Arabia’s Vision 2030 programme places weight on domestic digital capability and on keeping data generated in the Kingdom under national control. Sovereign edge AI addresses both directly: the inference happens on the operator’s own hardware, on their own site, with no requirement to send video streams to a third-party cloud in another jurisdiction.

The power argument matters just as much at scale. Continuous video analytics across hundreds of cameras is an always-on workload — the efficiency of the inference silicon sets the ceiling on how many streams a site can process within the power and cooling it actually has.