Beelink Launches Modular ME Pro Storage PCs for AI Workloads

A platform that evolves rather than a one-time purchase
Beelink positions the modular ME Pro as a long-term system where processor modules can be upgraded without replacing the entire chassis.

In the spring of 2026, Shenzhen-based Beelink introduced a line of compact computers that quietly redraws the boundary between personal workstation and storage infrastructure. The ME Pro series — offered in 2-bay and 4-bay configurations with Intel and AMD processors — can hold up to 136 terabytes of local data while running AI workloads without reliance on the cloud. More than a product launch, it reflects a broader reckoning in the technology industry: as creative and artificial intelligence work grows more data-hungry, the machines that support it must grow not just more powerful, but more enduring.

  • Studios and creative teams are drowning in data that consumer hardware was never built to hold, and the gap between what people need and what a standard desktop offers has become impossible to ignore.
  • Beelink's modular chassis design directly challenges the disposability cycle of modern computing — users can swap processor modules years from now without replacing the entire machine.
  • The flagship Ryzen AI 9 HX 370's 80 TOPS rating signals a push toward on-premise AI inference, keeping sensitive datasets and large language model processing off third-party cloud servers.
  • Dual 10GbE and 2.5GbE networking ports allow collaborative teams to synchronize massive video and audio assets at speeds that make real-time creative workflows across machines genuinely practical.
  • With 136TB capacity, ARM and color-variant editions on the horizon, and a platform framing rather than a fixed-product framing, Beelink is positioning the ME Pro as infrastructure that grows alongside its users.

Beelink unveiled new 2-bay and 4-bay configurations of its ME Pro compact computer line on May 29, 2026, built around Intel and AMD processors — including the Core i5-13420H, Ryzen 7 H 255, and the flagship Ryzen AI 9 HX 370. The machines are designed for the intersection of creative production and artificial intelligence, where local storage demands have outgrown what conventional desktops can offer.

What sets the ME Pro apart is its modular architecture. The drawer-style chassis accepts interchangeable processor modules, meaning users can upgrade the machine's core computing power in future years without replacing the storage infrastructure or starting over entirely. Intel models ship in navy blue, AMD versions in black with red accents, and an ARM-based edition in pearl white is already planned.

Storage capacity is the most dramatic departure from standard desktop computing. The 2-bay model reaches up to 76 terabytes across four SSD slots and two hard drive bays; the 4-bay version doubles that to 136 terabytes. These figures exist in a different category from the single drives measured in hundreds of gigabytes that ship in most consumer laptops — these machines are built for studios processing video, audio, and datasets that run into the dozens of terabytes.

The flagship HX 370 delivers up to 80 TOPS of AI computational throughput, enabling local inference and large language model processing without sending data to the cloud. Every configuration includes both a 10-gigabit and a 2.5-gigabit Ethernet port, supporting fast asset synchronization and advanced networking features across collaborative teams.

Beelink describes the ME Pro as an 'integrated AI storage-and-computing platform' — something that sits between a personal workstation and a network-attached storage system without being fully either. The modular design ensures the chassis remains relevant as processor technology advances, framing the purchase less as a fixed product and more as an evolving foundation.

Beelink has expanded its ME Pro line of compact computers into a platform designed for the moment when artificial intelligence and creative work collide with the need for enormous amounts of local storage. The company, based in Shenzhen, unveiled new 2-bay and 4-bay configurations on May 29, 2026, each built around processors from Intel and AMD—the Core i5-13420H, Ryzen 7 H 255, and the flagship Ryzen AI 9 HX 370. What distinguishes these machines is not just their power but their architecture: a modular design that lets users swap out the processor module without dismantling the entire system.

The ME Pro's drawer-style chassis and interchangeable motherboard modules represent a deliberate shift in how Beelink thinks about longevity. Rather than treating a computer as a fixed purchase, the company frames the ME Pro as an evolving platform. Future processor modules will slot into the same chassis, meaning a user who buys one today can upgrade the brain of the machine in two or three years without starting over. This modularity extends to aesthetics as well—Intel versions come in navy blue, AMD machines in black with red accents, and an upcoming ARM-based edition will arrive in pearl white.

The storage capacity is where these machines depart most sharply from conventional desktop computers. The 2-bay model combines four SSD slots with two hard drive bays, reaching up to 76 terabytes of total capacity. The 4-bay version doubles down, offering four SSDs and four HDDs for a maximum of 136 terabytes. For comparison, most consumer laptops ship with a single drive measured in hundreds of gigabytes. These machines are built for studios, collaborative teams, and anyone processing video, audio, or datasets that measure in the dozens of terabytes.

Performance matters too, particularly for the work these machines are meant to handle. The Ryzen 7 H 255 and Ryzen AI 9 HX 370 excel at multi-threaded tasks like editing video with multiple tracks and complex effects. The flagship HX 370 pushes further into artificial intelligence territory, delivering up to 80 TOPS—a measure of computational throughput—which positions it for local AI inference and processing large language models without sending data to the cloud. Every configuration includes dual Ethernet ports: a 10-gigabit connection and a 2.5-gigabit connection. This dual-network setup allows rapid synchronization of large creative assets between machines and enables advanced networking features like bandwidth aggregation.

What Beelink is describing amounts to a category that sits between a traditional personal computer and a network-attached storage system. The ME Pro is neither purely one nor the other. It is, in the company's framing, an "integrated AI storage-and-computing platform"—a machine that combines enough processing power to run demanding creative software and AI workloads with enough local storage to keep massive datasets on hand. The modular architecture ensures that as processor technology advances, the chassis and storage infrastructure remain relevant. Users are not locked into today's silicon; they are buying into a platform that evolves.

More than a PC or NAS, ME Pro is positioned as an integrated AI storage-and-computing platform where performance scales with user demand
— Beelink
A Conversa do Hearth Outra perspectiva sobre a história
Inventor

Why does a compact PC need 136 terabytes of storage? That's not a normal use case.

Model

It's not normal for a single person working alone. But a video studio with multiple editors, or a team training AI models locally, or someone processing large scientific datasets—they all need that density in a single, manageable box. Cloud storage has latency and bandwidth costs. This keeps everything local.

Inventor

The modular motherboard thing—is that actually useful, or is it marketing?

Model

It's genuinely useful if Beelink follows through. You buy the chassis and storage infrastructure once. In three years, when new processors arrive, you swap the module instead of replacing the whole machine. That's a real cost savings for studios that can't afford to refresh everything at once.

Inventor

What's the Ryzen AI 9 HX 370 actually good for? Eighty TOPS sounds impressive but I don't know what that means in practice.

Model

It means running AI models locally without sending your data to a cloud service. Video upscaling, image enhancement, voice processing—all on the machine itself. For creative professionals, that's privacy and speed. For enterprises, it's not having to pay per-API-call fees.

Inventor

So this is really a storage device that happens to have a powerful computer inside it?

Model

That's one way to see it. Or it's a powerful computer that happens to have massive storage built in. The point is they're integrated. You're not bolting a NAS onto a workstation. It's one unified thing.

Inventor

Who actually buys this? It seems niche.

Model

Video production studios, AI research teams, content creators working with 4K or 8K footage, collaborative teams that need shared local storage with compute power. Not niche in the sense of tiny—those are real, growing markets. But yes, not for someone checking email.

Quer a matéria completa? Leia o original em PR Newswire UK ↗
Fale Conosco FAQ