Skip to content

  • Storage powering real-time intelligence
    Storage powering real-time intelligence
    Storage powering real-time intelligence

    An image where streamlined graphics representing data flow converge, centered around the chip image of UFS 5.0, along with icons representing AI

Next-gen UFS 5.0 for edge AI
Next-gen UFS 5.0 for edge AI
Next-gen UFS 5.0 for edge AI

As AI workloads increasingly shift to the edge, devices must process data in real time while handling more complex and sensitive tasks locally. This demands a new class of storage with higher bandwidth, lower latency, and greater efficiency within constrained device environments. UFS 5.0 is built to meet these complicated requirements—providing an optimized foundation for next-generation edge AI devices. As AI workloads increasingly shift to the edge, devices must process data in real time while handling more complex and sensitive tasks locally. This demands a new class of storage with higher bandwidth, lower latency, and greater efficiency within constrained device environments. UFS 5.0 is built to meet these complicated requirements—providing an optimized foundation for next-generation edge AI devices. As AI workloads increasingly shift to the edge, devices must process data in real time while handling more complex and sensitive tasks locally. This demands a new class of storage with higher bandwidth, lower latency, and greater efficiency within constrained device environments. UFS 5.0 is built to meet these complicated requirements—providing an optimized foundation for next-generation edge AI devices.

Boost edge AI performance Boost edge AI performance Boost edge AI performance

Edge AI’s growth requires mobile devices to access large, scattered data in real time.

UFS 5.0—the next-generation UFS standard built on UniPro 3.0 and M-PHY 6.0—boosts AI performance with up to 10.8 GB/s of bandwidth, delivering 2x faster sequential read/write speeds, and 5x faster random read speeds than the previous generation1). By leveraging optimized PAM4 signaling and an efficient protocol architecture that reduces overhead, it supports low‑latency, responsive AI across mobile devices, wearables, automotive systems, and next-generation devices.


1) Compared to UFS 4.1

Edge AI’s growth requires mobile devices to access large, scattered data in real time. UFS 5.0—the next-generation UFS standard built on UniPro 3.0 and M-PHY 6.0—boosts AI performance with up to 10.8 GB/s of bandwidth, delivering 2x faster sequential read/write speeds, and 5x faster random read speeds than the previous generation1). By leveraging optimized PAM4 signaling and an efficient protocol architecture that reduces overhead, it supports low‑latency, responsive AI across mobile devices, wearables, automotive systems, and next-generation devices.


1) Compared to UFS 4.1

Edge AI’s growth requires mobile devices to access large, scattered data in real time. UFS 5.0—the next-generation UFS standard built on UniPro 3.0 and M-PHY 6.0—boosts AI performance with up to 10.8 GB/s of bandwidth, delivering 2x faster sequential read/write speeds, and 5x faster random read speeds than the previous generation1). By leveraging optimized PAM4 signaling and an efficient protocol architecture that reduces overhead, it supports low‑latency, responsive AI across mobile devices, wearables, automotive systems, and next-generation devices.


1) Compared to UFS 4.1

Linear graphics representing data flow and AI icons flow rapidly over a mobile device, and above them, a UFS 5.0 chip image appears along with the bandwidth text 10.8 GB/s being highlighted.
Optimized power efficiency for mobile devices Optimized power efficiency for mobile devices Optimized power efficiency for mobile devices

Delivering higher performance no longer requires higher power consumption. UFS 5.0 offers a 40 % improvement in power efficiency over the previous generation2) through optimized signaling, improved interface efficiency, and enhanced power integrity. This reduces unnecessary energy use during data transfer, enabling mobile and wearable devices to optimize power efficiency, extend battery life, and handle demanding AI workloads.


2) Compared to UFS 4.1 under the same sequential read performance conditions

Delivering higher performance no longer requires higher power consumption. UFS 5.0 offers a 40 % improvement in power efficiency over the previous generation2) through optimized signaling, improved interface efficiency, and enhanced power integrity. This reduces unnecessary energy use during data transfer, enabling mobile and wearable devices to optimize power efficiency, extend battery life, and handle demanding AI workloads.


2) Compared to UFS 4.1 under the same sequential read performance conditions

Delivering higher performance no longer requires higher power consumption. UFS 5.0 offers a 40 % improvement in power efficiency over the previous generation2) through optimized signaling, improved interface efficiency, and enhanced power integrity. This reduces unnecessary energy use during data transfer, enabling mobile and wearable devices to optimize power efficiency, extend battery life, and handle demanding AI workloads.


2) Compared to UFS 4.1 under the same sequential read performance conditions

40% Improved power efficiency letters graphically highlighted next to a UFS 5.0 chip image on a mobile device
Compact, yet full capacity Compact, yet full capacity Compact,
yet full capacity

UFS 5.0 provides up to 1 TB of storage capacity in a compact 7.5x13.0 mm package, enabling mobile devices to store enough data to perform their own AI inference. This size optimization improves space efficiency without compromising capacity. The smaller form factor allows more flexible device design, freeing up internal space for critical components such as batteries, making it ideal for space-constrained applications, including smartphones or wearable devices.

UFS 5.0 provides up to 1 TB of storage capacity in a compact 7.5x13.0 mm package, enabling mobile devices to store enough data to perform their own AI inference. This size optimization improves space efficiency without compromising capacity. The smaller form factor allows more flexible device design, freeing up internal space for critical components such as batteries, making it ideal for space-constrained applications, including smartphones or wearable devices.

UFS 5.0 provides up to 1 TB of storage capacity in a compact 7.5x13.0 mm package, enabling mobile devices to store enough data to perform their own AI inference. This size optimization improves space efficiency without compromising capacity. The smaller form factor allows more flexible device design, freeing up internal space for critical components such as batteries, making it ideal for space-constrained applications, including smartphones or wearable devices.

Graphics emphasizing the reduction of the UFS 5.0 form factor on a mobile mainboard, with "16.7% smaller size" highlighted in text on the left and "up to 1TB" specs highlighted in text on the right

Applications for UFS 5.0

Explore more about UFS 5.0

* 1 TB = 1,000,000,000,000 Bytes, 1 GB = 1,000,000,000 Bytes. Actual usable capacity may be less (due to formatting, partitioning, operating system, applications or otherwise).
* NVM Express® design mark and NVMe® word mark are trademarks of NVM Express, Inc.
* PCI Express® and PCIe® are registered trademarks of PCI SIG.
* The contents of this page are provided for informational purposes only. No representation or warranty (whether express or implied) is made by Samsung or any of its officers, advisers, agents, or employees as to the accuracy, reasonableness or completeness of the information, statements, opinions, or matters contained in this page, and they are provided on an "AS IS" basis.
* Samsung will not be responsible for any damages arising out of the use of, or otherwise relating to, the contents of this page. Nothing in this page grants you any license or rights in or to information, materials, or contents provided in this document, or any other intellectual property.
* The contents of this page may also include forward looking statements. Forward looking statements are not guarantees of future performance and that the actual developments of Samsung, the market, or the industry in which Samsung operates may differ materially from those made or suggested by the forward looking statements contained in this page.
* All data on products herein, including their performances, are based on internal testing using standard Samsung benchmarks under laboratory conditions.
* All design, features and specifications represented herein may change without notice.
* Images shown here have been adjusted for demonstration purposes and may appear differently on the actual products.
* Test results do not guarantee future performance under such test conditions, and the actual throughput or performance that any user will experience may vary depending upon many factors.
* For further details on product specifications, please contact the sales representative of your region.

https://semiconductor.samsung.com/support/contact-info/global-network/