HP caps off its PC overhaul with the launch of the OmniBook Ultra 14 – its most powerful AI-powered laptop to date

HP OmniBook Ultra 14 concept image showing laptop on a custom background with black and yellow coloring.
(Image credit: Future/HP)

HP has launched the most powerful AI PC available on the market, the OmniBook Ultra 14, as it revitalizes its computer portfolio ahead of what it expects to be an oncoming wave of demand for AI compute at the edge. 

Unveiled at HP’s Imagine AI event in New York, Alex Cho, president of personal systems, touted the device as the “most powerful AI PC in the industry”, with an AMD-based neural processing unit (NPU) that boasts an impressive 55 trillion operations per second (TOPS).

Cho said the OmniBook 14’s AMD-based chip notches a 45% increase in compute power over Apple’s M4 chip, the current benchmark for mobile processors, and twice the performance of the previous M3,

Notably, 55 TOPS would surpass the peak performance levels AMD said was achievable with its Ryzen AI 300 series processors, which Cho explained was the result of a special partnership with the chip manufacturer.

“We are doing a lot more curation of the stack. I said that back in Palo Alto, we are curating the stack from silicon, bios, OS, applications, IO, etc. and doing that deep coengineering with [AMD] allowed us to deliver a level of performance that is above the rest of the industry,” he said. 

The OmniBook will head up HP’s consumer product line, while its flagship enterprise device, the EliteBook Ultra G1q was launched at Microsoft’s Surface Event in May.

The EliteBook will sport Qualcomm’s latest 12-core Snapdragon X Elite processor, which features a 45 TOPS NPU, although Alex Thatcher, senior director of cloud client and AI experience at HP, told ITPro this number edges closer to 75 TOPS when you include the GPU.

Both these devices will also look to challenge the MacBook lineup’s stranglehold on the best battery life of any laptop. By allowing AI workloads to be offloaded to the high efficiency NPU, HP can free up the CPU and GPU for other tasks, with HP claiming the EliteBook Ultra will offer 26 hours of run time.

OmniBook Ultra 14 launch marks the next step in HP’s device refresh 

HP announced a total rebrand of its entire PC lineup in May, doing away with its Pavilion, Dragonfly, Spectre, and Envy models in favor of a simplified roster of products targeting consumers and enterprise customers.

Its new PC offering is split into two product lines, Omni devices for consumers and Elite products for businesses. Laptops in these lineups will be known as either OmniBooks or EliteBooks, while all-in-one systems have become OmniStudio and Elite Studio. Finally, traditional desktop PCs  will now be OmniDesk and EliteDesk.

For each of these product categories, HP will offer a number of different tiers which cater to different needs and price points. The Ultra and X labels are reserved for the more premium devices with the most high-end components. 

From there, each line will have a number of different skews denoted by odd numbers for the consumer side, and even numbers for the commercial portfolio.

Speaking to ITPro, Guayente Sanmartin, SVP & division president of commercial systems at HP, outlined the reasoning behind this overhaul, stating the goal was to streamline its device ecosystem to better reflect HP’s core values and mark what it sees as a new era in personal computing.

“What we have done is simplified the branding as a way to ensure that the HP values were properly communicated…and the objective was starting this new era where HP has a lot of AI agility,” she said. 

Expanding on these values, Sanmartin said first among these is reliability, noting HP is primarily a brand built on trust, and that within its sub brands it tries to communicate slightly different values depending on the personas they are trying to serve.

In terms of the commercial lineup, which Sanmartin heads up, she explained consistency was central to their value proposition. Citing Alex Cho’s comments earlier that day, she reasserted that no one spends more time with an employee from a device perspective than HP, and so getting these experiences right is imperative for the company.

HP is bullish on the value AI PCs can unlock for businesses

This overhaul is a declaration of intent from HP, indicating it’s bullish on widespread adoption of AI PCs moving forward, as software vendors catch up on building tools that can leverage these local AI accelerators and end users get a better understanding of the value they provide.

Sanmartin said she expects an explosion of demand for AI computing at the edge over the next few years, noting HP’s projections indicate that within three years time 50% of all PCs will be AI PCs, and HP wants to be at the center of this new market.

Explaining where HP expects to separate its AI PC lineup from its competitors like Dell and Lenovo, is by focussing on the user and maximizing the impact of the use cases these devices will serve, highlighting three pillars by which it hopes to show it is leading the field.

“I think we are going to be differentiated in performance, differentiated in our capacity to personalize the device more… and the third thing is privacy and security… we are a trust partner, a partner you can trust on ethics, sustainability, all of this is our differentiation”,

RELATED WHITEPAPER

Providing an example of this approach, Thatcher noted the company’s decision to take the time to totally redesign the EliteBook Ultra, instead of taking the easier path of building on an existing chassis.

“When we did our Qualcomm device we didn’t just take an existing chassis and put a board in it we did a completely ground up design, and the benefit of that is you see it in the battery life. That’s why we get 26 hours of battery life.”

Solomon Klappholz
Staff Writer

Solomon Klappholz is a Staff Writer at ITPro. He has experience writing about the technologies that facilitate industrial manufacturing which led to him developing a particular interest in IT regulation, industrial infrastructure applications, and machine learning.