Disclosure: Some links on this page are monetized by the Skimlinks, Amazon, Rakuten Advertising, and eBay, affiliate programs, and Liliputing may earn a commission if you make a purchase after clicking on those links. All prices are subject to change, and this article only reflects the prices available at time of publication.

Network Attached Storage company QNAP has unveiled a system that it’s positioning as its cheapest dual-bay NAS with a 2.5 GbE port for high-speed data transfer over a local network. The new QNAP TS-216G is also likely to be the company’s cheapest dual-bay model with an integrated NPU (Neural Processing Unit) for hardware-accelerated AI features including image recognition.

AI is all the rage in the consumer PC and mobile spaces at the moment, so it was probably just a matter of time until we started seeing companies tout the AI features of storage devices… but in this case it might not just be about hype.

The TS-216G features a 2 GHz ARM Cortex-A55 quad-core processor, Mali-G52 graphics, and an unspecified neural processing unit. But if I had to guess, I’d say it’s probably a Rockchip RK3566 processor with a 0.8 TOPS NPU, as that’s the chip QNAP already uses for the TS-233, which launched about two year ago.

But the TS-233 is a ships with only 2GB of RAM and a single Gigabit Ethernet port. It’s positioned as an entry-level network-attached storage solution.

QNAP’s new dual-bay model allows you to add redundant (and/or additional) storage, depending on how you configure your hard drives. It ships with 4GB of RAM. And it has two RJ45 Ethernet jacks: one with support for 2.5 GbE speeds and a second that supports Gigabit Ethernet.

So what are you supposed to actually use the NPU for? At this point it seems like image recognition is the main use case.

QNAP Photo Station software basically lets you set up your own private alternative to Google Photos or iCloud Photos by setting up a “private cloud” service that lets you view your images on any device with an internet connection. The NPU can assist in identifying people, places, and things in the photos to help you sort or search your photos more quickly – and QNAP says the NPU in the TS-216G speeds up facial recognition by 600% while consuming less power.

The company also offers subscription-based surveillance software that can leverage the NPU for more efficient face recognition.

The TS-216G measures 8.7″ x 6.5″ x 4″ and weighs 3.2 pounds. It supports up to two 3.5 inch or 2.5 inch SATA 3 hard drives and consumes around 5 to 14 watts of power (5 watts on average when the hard drives are sleeping and 14 watts during active use). There’s a 120mm fan inside the case, so don’t expect the system to be entirely silent, but at those power levels, it’s likely that the fan won’t have to run constantly.

It has one USB 3.2 Gen 1 Type-A port and two USB 2.0 ports. One thing to keep in mind is that the 4GB of RAM included is all this system will ever support: there’s no way to upgrade the system’s memory.

QNAP hasn’t announced pricing for the TS-216G yet, btu the company is positioning it as a “mid-range” solution that falls somewhere between the TS-233 ($199) and the TS-262 ($369), which has an Intel Celeron N4505 processor, two M.2 2280 slots for optional SSDs, and a PCIe Gen 3 x2 slot.

press release

Support Liliputing

Liliputing's primary sources of revenue are advertising and affiliate links (if you click the "Shop" button at the top of the page and buy something on Amazon, for example, we'll get a small commission).

But there are several ways you can support the site directly even if you're using an ad blocker* and hate online shopping.

Contribute to our Patreon campaign


Contribute via PayPal

* If you are using an ad blocker like uBlock Origin and seeing a pop-up message at the bottom of the screen, we have a guide that may help you disable it.

Subscribe to Liliputing via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 9,546 other subscribers

Join the Conversation


Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

  1. Oh, do people expect me to be tagging all my images now? Well, there’s this thing and DigiKam 8.3 (only works on faces for now), but I have no idea how well these work.
    I do kind of want that feature, but what I really want it to do, is to tag images based on the arguments I might use them as evidence in. For now I’ve been getting along with my directory structure and typing in filenames that contain every word I might search for later.

  2. There are no neurons in an NPU and no retinas in a “Retina” display. I wish the tech industry would have the restraint to not use such magical terminology.