Close Menu
    Sultanate TimesSultanate Times
    • Automotive
    • Business
    • Entertainment
    • Health
    • Lifestyle
    • Luxury
    • News
    • Sports
    • Technology
    • Travel
    Sultanate TimesSultanate Times
    Home » Think unveils Think Fabric™, a new approach to bare-metal, heterogeneous AI compute built for the age of efficiency
    PR Newswire

    Think unveils Think Fabric™, a new approach to bare-metal, heterogeneous AI compute built for the age of efficiency

    August 31, 2026
    Facebook WhatsApp Twitter Pinterest LinkedIn Telegram Tumblr Email Reddit VKontakte
    • New platform brings together Think’s three core technologies: Node, ILM and Constellation into a single, integrated ultra high-efficiency AI infrastructure solution
    • Think Fabric is the first bare-metal AI compute platform to enable heterogeneous high-performance AI compute with mixed silicon, providing greater efficiency, flexibility and control
    • Think Fabric is on show at LEAP 2026 as part of Saudi Arabia’s next generation of AI infrastructure innovation

    RIYADH, Saudi Arabia, Aug. 31, 2026 /PRNewswire/ — Think, the AI infrastructure company building for the age of efficiency, today announced Think Fabric, a new category of unified AI infrastructure that combines intelligent software with high-performance hardware. Think Fabric is designed to address some of the most significant efficiency bottlenecks in AI infrastructure today, including cooling, density, GPU utilisation and infrastructure cost, enabling enterprises, governments and AI-native organisations to deploy high-performance AI with greater efficiency, flexibility and control.

    Think Logo

    Think Fabric brings together Think’s three core technologies into a single, tightly integrated platform:

    • Node is Think’s answer to the challenge of delivering high-density AI compute in a compact form factor without the need for extensive cooling infrastructure. Despite having the same GPU count and silicon as a typical AI data centre server, Node delivers radically higher sustained compute and dramatically higher sustained power density, enabled by Think’s patent-pending sealed-loop cooling. Each GPU uses a patent-pending high-density cooling topology to more efficiently extract heat from the GPU core and memory, enabling sustained, high-density AI compute in a form factor with up to 2.35x greater power density than any other server manufacturer with like-for-like specifications.
    • ILM (Intelligent LLM Management system) is Think’s proprietary software orchestration layer designed to dramatically improve GPU utilisation. ILM intelligently optimises how workloads are distributed across AI accelerators, reducing the wasted energy and cost associated with expensive idle silicon operating below capacity. ILM can run multiple AI models on a single GPU, allowing multiple models to share GPUs that would otherwise sit idle, substantially reducing the total hardware required.
    • Constellation links multiple Think Nodes into a unified supercompute cluster using a high-bandwidth fibre optic interconnect for low-latency, multi-node coordination, with ILM handling synchronisation and local orchestration. Constellation can also enable organisations to scale high-performance AI compute clusters without the footprint and complexity associated with traditional data centre infrastructure.

    Think Fabric is being demonstrated publicly for the first time at LEAP 2026 in Riyadh.

    “This first era of AI has been defined by the rush to scale, but it’s clear that the long-term success of AI will only happen by addressing the issues around efficiency and cost,” said Ahmed AlSharif, Founder and CEO of Think. “Think Fabric was built around a simple principle: we should be able to do more with the compute we already have. The solution to the rising cost of AI infrastructure is not to spend even more or build bigger – it’s to maximise the investments already made, and to lower the cost of AI compute through increased efficiency.”

    “AI adoption shouldn’t require organisations to compromise on sovereignty, economics or flexibility,” said Ammar Enaya, Co-Founder and COO of Think. “Fabric gives organisations a platform they can own, control and evolve over time, without sacrificing performance or operational simplicity.”

    Think believes the billions already invested in AI compute capacity, the rising cost of essential components such as memory, and the growing power and environmental demands of data centres are pushing the industry towards a new phase in which the focus shifts from simply adding more infrastructure to extracting greater value from the compute already available.

    “We’re moving into a world where efficiency matters more than raw scale,” added AlSharif. “The future of AI infrastructure will be defined by how much performance we can get from existing investments. The industry needs to shift its focus from how much raw compute it can build to the metrics that really matter: ROI and the cost per token of AI.”

    Think Fabric forms the foundation of Think’s approach to the age of efficiency, bringing hardware, cooling, orchestration and interconnect together as a unified system designed to help organisations get more from every unit of AI compute.

    About Think
    Think is a Saudi AI infrastructure company building the engine room for the age of efficiency. Its unified approach brings together intelligent software and high-performance hardware to make AI compute more efficient, accessible and economically sustainable.

    Think’s technology addresses the fundamental bottlenecks limiting AI infrastructure today, including GPU utilisation, memory, cooling, power and cost. Through its Think Fabric infrastructure platform and Think Grid compute service, Think enables enterprises, governments and AI-native organisations to deploy high-performance AI with greater efficiency, flexibility and control.

    Founded by Ahmed AlSharif, a technology leader whose experience spans Meta, PlayStation, and EA, and enterprise technology veteran Ammar Enaya, formerly of Cisco, Aruba, and HPE, Think is headquartered in Riyadh, Saudi Arabia.

    Find out more at www.think-ai.com and https://www.linkedin.com/company/think-ai-labs/

     

    Cision View original content to download multimedia:https://www.prnewswire.co.uk/news-releases/think-unveils-think-fabric-a-new-approach-to-bare-metal-heterogeneous-ai-compute-built-for-the-age-of-efficiency-302864569.html

    Related Posts

    Can Doing Good Create Business Value?

    September 10, 2026

    Aircom Appoints Khurram Chaudhry as General Manager

    September 10, 2026

    RAMINVEST HOLDING Advances Mauritania Gold Exploration Following 50.14 g/t Gold Assays at Wompou Project

    September 10, 2026

    World’s First Robot Sports League Launched in Riyadh, Plans to Hit 8 Cities Worldwide

    September 9, 2026

    $1 Million Awarded to Social Impact Student Start-Up at 2026 EF Hult Prize

    September 9, 2026

    Adani Airports to raise ~USD 1 billion of primary equity from marquee global investors

    September 9, 2026
    Latest News

    UAE-Germany ties in focus during presidential visit

    September 10, 2026

    iPhone 18 Pro brings variable aperture and A20 Pro chip

    September 10, 2026

    Apple launches iPhone Duo foldable starting at $1,999

    September 10, 2026

    Gold steadies near $4,400 with US inflation in focus

    September 10, 2026

    Ballon d’Or 2026 shortlist sets stage for London awards

    September 9, 2026

    US battery growth still depends on Chinese materials

    September 9, 2026

    Panama Canal could cut daily ship limit to 29

    September 9, 2026

    Etihad sets October start for Abu Dhabi Saudi Red Sea route

    September 9, 2026
    © 2026 Sultanate Times | All Rights Reserved
    • Home
    • Contact Us

    Type above and press Enter to search. Press Esc to cancel.