Intel Trades On The Versatility Of The Arc Pro Chip Series

Although Nvidia absolutely dominates GPUs, it's sometimes useful to look at some of the also-rans that take up parts of the market. There's interesting news from intel, where makers are unveiling a set of chips that can outfit various AI projects with the efficiency of smaller but still agile systems that aren't so much "Cadillac" options.

First, the top-shelf model, the Arc Pro B70, has 32 Xe2-HPG cores, with 32 ray-tracing units and 256 XMX engines for performance of around 367 TOPS. This design also features an upgraded 256-bit memory controller and expanded memory support, with 32GB of GDDR6 RAM and a rated bandwidth of 608GBps.

It's an AI GPU – not a gaming card. But you wouldn't know it from the naming of the architecture built in, where Intel referred to its first-gen pass as "Alchemist" or Xe1, and the subsequent Battlemage or Xe2 technology representing round two, for higher TOPS and better AI/upscaling support.

Anyway, the more mature design of Xe2, along with other selling points of the Arc Pro series, means developers are paying attention.

Going Strong on Memory

With so much RAM in the Intel chips, some might suggest the timing of this move is questionable.

"The increase in memory capacity may seem questionable at this time, given the ongoing memory shortage, but the large memory capacity is one of the most alluring features of these GPUs," writes Michael Justin Allen Sexton at PCMag. "While virtually everything in your computer can benefit from more and faster memory, AI workloads are particularly greedy on that front, and the local RAM allocation limits the size of models you can host in memory. The more complex the AI workload is, the more RAM is required to host and run the AI model, which makes memory capacity of paramount importance. The wider 256-bit memory interface also boosts memory bandwidth, which will improve overall performance."

Sexton also refers to an advantage of the memory-heavy build in implementations.

"Intel's internal testing showed that the Arc Pro B70 achieved significantly higher token throughput while handling AI workload requests from multiple users," Sexton writes. "At the same time, Intel also showed that the Arc Pro B70 would be a superior option for running large AI models due to its higher memory capacity."

As for parallel deployments, Sexton writes:

"Intel notes that these cards can be set up in arrays of two, four, or eight cards in a workstation or server rack module, to enable multiple AI models queued up in one workstation as needed, or to pool memory allowances for AI models that are too big for one card's memory allocation."

Playing Games

All of this power packed into the Arc Pro B70, testers estimate , leads to a 45% improvement in rendering for games like Cyberpunk 2077, Monster Hunter Wilds, and Shadow of the Tomb Raider.

But these chips are also good for AI.

"For anyone who has been following Intel's GPU journey, the Arc Pro B70 is a big deal," writes Rachit Agarwal for Digital Trends. "The Arc Pro B70 is the answer to anyone asking Intel to create an ultra-powerful GPU to power their workstation."

And then there's cost.

"Priced at $949, the Arc Pro B70 is drastically less expensive than the $1,800 RTX Pro 4000 from Nvidia and the $1,299 Radeon AI Pro R9700 from AMD," writes Timothy Green for Motley Fool . "While Nvidia retains an important advantage with CUDA, its proprietary software platform that protects its massive market share, Intel's claimed performance advantage could be enough for the company to make some inroads in this market."

Chart Moves

The news also seems to be spiking Intel's stock: the share price is up about 20% over a couple of weeks, at one-month highs, and very near all-time highs, given a top peak of around $54 per share in January. The story that the chart is telling is one of Intel becoming a bigger competitor in the market.

In terms of custom board partner designs, it looks like Maxsun's Arc Pro B70 powered cards are going to be popular, with fanless designs that have a lot of promise for the rack.

So it might make sense to keep an eye on this company as you consider what purchasers are going to be doing to support that next round of AI implementations in Q2 of 2026. Stay tuned for more.

This article was originally published on Forbes.com