The Scrappy $949 GPU Taking Aim at Nvidia's Grip on Local AI
Key Points
The Arc Pro B70 comes with 32GB or RAM, enabling smaller AI models to run locally.
It compares favorably with products from Nvidia and AMD, and it's much cheaper at $949.
If AI workloads move away from mega data centers, Intel is well positioned to benefit.
Intel (NASDAQ: INTC) just announced a new GPU. It's not for gaming, and it's not destined for data centers. Instead, the Intel Arc Pro B70 is tailor-made for running AI workloads on workstations locally.
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. 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.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
Image source: Intel.
Beating Nvidia at its own game
Frontier AI models require enormous quantities of memory and computing power, so they can only be run in massive data centers with high-powered AI accelerators. However, not every AI workload requires Claude Opus or GPT-5.4. There are plenty of use cases that can be satisfied with smaller AI models that fit on a workstation GPU.
The Arc Pro B70 has 32 Xe cores that deliver 22.9 TFLOPS of FP32 compute performance. It features 32GB of GDDR6 RAM, enabling 608 GB/s bandwidth. This type of memory is less performant than the high-bandwidth memory typically used in data center GPUs, but it's also cheaper. Intel claims the Arc Pro B70 can achieve 367 TOPS (trillions of operations per second).
The high memory content enables running a wider range of AI models with larger contexts. Using the open-source Llama 3.1 8B model, Intel claims that the Arc Pro B70 can handle a context window more than twice as large as the Nvidia RTX Pro 4000. Intel also claims that its new GPU delivers drastically faster responses, even with many concurrent users. Multiple GPUs can be linked together, enabling even larger context windows for larger AI models that can't fit on a single GPU.
Intel is positioning the Arc Pro B70 as a strong value compared to the alternatives. Compared to the RTX Pro 4000, the Arc Pro B70 can deliver up to two times as many tokens per dollar, depending on the model. The Arc Pro B70 is also aimed at users running professional or creative applications.
The future of AI could be local
It seems inevitable that AI workloads will eventually shift away from mega data centers. For enterprise AI inference applications with predictable loads that can use smaller AI models, GPUs running in on-premises or private cloud environments would be cost-effective, and data wouldn't be sent to the cloud.
Longer term, workstations and later standard PCs will eventually have enough memory and AI computing capacity to handle more sophisticated workloads. Imagine something like Claude Cowork, which can automate processes on a PC, running entirely locally. Not only does that solve privacy issues, but it also removes monthly subscription costs for users.
Intel appears well-positioned to be at the center of any shift toward local AI. Not only does the Arc Pro B70 push the envelope and enable more AI workloads to run on powerful workstations, but the company has also been advancing the AI capabilities of its latest PC chips. The top-tier Panther Lake laptop chip, for example, can deliver 180 TOPS of performance via its built-in AI accelerator and integrated graphics.
The ongoing memory shortage is a severe headwind right now, and prices for memory chips will need to come down before PCs have enough memory to support more capable AI models. The fact that Intel was able to squeeze 32GB of memory into the Arc Pro B70 is impressive, given how quickly memory prices have escalated.
The AI industry is focused on massive data center projects right now, and Intel is looking to play in that market as well with its upcoming Crescent Island GPU. But with the Arc Pro B70, Intel has shown a glimpse of what may be possible down the road as more AI computing power and memory find their way into workstations and PCs.
Should you buy stock in Intel right now?
Before you buy stock in Intel, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Intel wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $497,659!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,095,404!*
Now, it’s worth noting Stock Advisor’s total average return is 912% — a market-crushing outperformance compared to 185% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
*Stock Advisor returns as of March 27, 2026.
Timothy Green has positions in Intel. The Motley Fool has positions in and recommends Advanced Micro Devices, Intel, and Nvidia. The Motley Fool has a disclosure policy.
