- Thunderbolt 5 bandwidth pushes external GPU hardware closer to the workstation realm
- Local AI inference is gaining attention as cloud costs continue to rise
- Developers are increasingly exploring running language models directly on personal hardware
External GPU enclosures have been around for some time – typically associated with gaming laptops and graphics acceleration tasks that exceed the capabilities of mobile processors.
Plugable’s recently released TBT5-AI falls into this category, but introduces a design focused on connecting desktop graphics hardware to laptops for local AI workloads.
The enclosure provides a full-length PCIe x16 slot that allows users to install a desktop-class graphics card inside the external enclosure.
The article continues below
Desktop-class hardware in an external enclosure
An integrated 850-watt power supply provides the energy required to run high-performance GPUs that would normally only function inside desktop workstations.
For connectivity, this device comes with a single Thunderbolt 5 cable, which allows direct connection with a laptop and supports up to 80 Gbps of bi-directional bandwidth, while a boost mode can increase throughput to 120 Gbps for certain workloads.
Inside the case, this bandwidth connects the installed GPU through PCIe 4.0 x4 lanes, reducing the transfer bottlenecks that limited previous external GPU designs.
In addition to housing the graphics card, the system acts as a hub that extends the connection to the connected laptop.
It delivers up to 96 watts of charging power, while also providing 2.5-gigabit Ethernet networking and multiple high-speed USB ports.
According to Plugable, many engineers increasingly want to keep model processing and data management in their own systems, and TBT5-AI offers just that as it is designed for developers experimenting with local AI inference environments.
The device allows developers to run large language models directly on local hardware instead of sending workloads to cloud infrastructure.
It supports common local AI frameworks including llama.cpp, Hugging Face models and Nvidia’s NIM inference platform.
Plugable chief technology officer Bernie Thompson said the hardware is aimed at industries where protecting sensitive information remains a strict operational requirement.
“Data protection is not a function, but a mandate,” Thompson said, referring to sectors such as health care, financial services and legal organizations.
Plugable is also preparing enterprise versions called TBT5-AI16, TBT5-AI32 and TBT5-AI96 that will include bundled graphics processors.
These configurations will integrate a software environment called Plugable Chat, described as an air-gapped AI orchestration platform for regulated organizations.
The company claims these systems will move AI processing away from subscription-based cloud services to locally managed computing infrastructure.
Priced at $599.95 as a standalone unit, the Plugable TBT5-AI enclosure was officially released a few days ago and is now available via Amazon and Plugable.com.
Via Macsource
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds. Be sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, video unboxings, and get regular updates from us on WhatsApp also.
