Argyll Data Development launches fully sovereign AI inference cloud

New sovereign UK inference cloud is made possible by energy-efficient SambaNova AI infrastructure.

  • Tuesday, 12th May 2026 Posted 58 minutes ago in by Phil Alsop

Argyll Data Development has launched its sovereign AI inference cloud, built in partnership with SambaNova, giving UK organisations a way to run production AI workloads without relying on foreign-owned hyperscale infrastructure. 

 

As enterprises move AI from pilots into live environments, questions around data sovereignty, regulatory exposure and infrastructure cost are becoming critical. While many organisations are currently investing heavily in GPU-based deployments, these approaches often come with significant power demands, complex cooling requirements and ongoing cost challenges, alongside continued reliance on cloud inference providers where control ultimately sits outside the UK. 

 

Argyll’s platform addresses this gap by combining UK-owned infrastructure with SambaNova’s advanced AI hardware and software stack, ensuring all data, models and operations remain fully under UK control. 

 

At the core of the platform is SambaNova’s Reconfigurable Data Unit (RDU) architecture running its SambaManaged solution. This gives customers high-performance AI inference without the additional power and exotic cooling demands associated with traditional GPU-based systems. Operating within a power envelope of approximately 10kW per rack, SambaNova's technology allows sovereign AI infrastructure to be deployed within existing UK data centres. 

 

The efficiency of Argyll's inference cloud enables organisations to run AI applications in real time - from customer operations to fraud detection - at a cost profile that makes production deployment viable. It hosts the latest open source models, such as Minimax running at speeds of up to 400 tokens per second, delivered entirely within a UK-resident environment. 

 

Peter Griffiths, Chairman at Argyll Data Development, said: 

 "Sovereignty in AI is not a label you can apply to a contract or a colocation agreement. It is a condition that has to be demonstrated - who is accountable, where the infrastructure sits, who controls the intelligence layer, and whether all of that aligns with the expectations of the society being served. Our platform satisfies those conditions. We are building the standard that others should be measured against." 

 

 

Jude Sheeran, Managing Director, EMEA at SambaNova added: 

 "As organisations scale AI, many are defaulting to GPU infrastructure without fully accounting for long-term cost, energy and operational complexity. Our work with Argyll provides an alternative, enabling high-performance AI inference that is more efficient, deployable and aligned with sovereignty requirements." 

 

Argyll’s inference cloud has been architected as a disaggregated platform, allowing compute, storage and networking to be deployed across multiple UK locations while operating as a single, unified inference layer. This provides added resilience and flexibility for organisations operating in regulated or security-sensitive sectors. 

An examination of how Atlassian’s Rovo and Teamwork Graph introduce AI-driven automation into...
Two thirds of organisations (64 per cent) are actively using artificial intelligence across the UK,...
Keeper Security has released its latest global insight report, “Identity Security at Machine...
WatchGuard® Technologies says that it has redefined how managed service providers (MSPs) deliver...
Study finds most organizations recognize the need for connected data, content, and workflows, but...
A third (35%) of European organisations cannot say whether they have been hit by an AI-powered...
Atlassian has announced a series of updates across its platform, including expanded capabilities...
Powered by Kaseya Intelligence, Kaseya’s platform said to combine the industry’s deepest...