- AMD mentioned its forthcoming most-advanced GPU for synthetic intelligence, the MI300X, will begin transport to some prospects later this 12 months, the corporate mentioned on Tuesday.
- AMD’s announcement on Tuesday represents the strongest problem to Nvidia, which presently dominates the marketplace for AI chips.
Lisa Su shows an ADM Intuition M1300 chip as she delivers a keynote tackle at CES 2023 at The Venetian Las Vegas on January 04, 2023 in Las Vegas, Nevada.
David Becker | Getty Pictures
AMD mentioned on Tuesday its most-advanced GPU for synthetic intelligence, the MI300X, will begin transport to some prospects later this 12 months.
AMD’s announcement represents the strongest problem to Nvidia, which presently dominates the marketplace for AI chips with over 80% market share, in keeping with analysts.
GPUs are chips utilized by corporations like OpenAI to construct cutting-edge AI packages comparable to ChatGPT.
If AMD’s AI chips, which it calls “accelerators,” are embraced by builders and server makers as substitutes for Nvidia’s merchandise, it might signify an enormous untapped marketplace for the chipmaker, which is greatest recognized for its conventional pc processors.
AMD CEO Lisa Su advised buyers and analysts in San Francisco on Tuesday that AI is the corporate’s “largest and most strategic long-term progress alternative.”
“We take into consideration the info heart AI accelerator [market] rising from one thing like $30 billion this 12 months, at over 50% compound annual progress fee, to over $150 billion in 2027,” Su mentioned.
Whereas AMD did not disclose a worth, the transfer might put worth stress on Nvidia’s GPUs, such because the H100, which might value $30,000 or extra. Decrease GPU costs could assist drive down the excessive value of serving generative AI purposes.
AI chips are one of many vibrant spots within the semiconductor business, whereas PC gross sales, a standard driver of semiconductor processor gross sales, hunch.
Final month, AMD CEO Lisa Su mentioned on an earnings name that whereas the MI300X will likely be obtainable for sampling this fall, it could begin transport in better volumes subsequent 12 months. Su shared extra particulars on the chip throughout her presentation on Tuesday.
“I like this chip,” Su mentioned.
AMD mentioned that its new MI300X chip and its CDNA structure had been designed for giant language fashions and different cutting-edge AI fashions.
“On the heart of this are GPUs. GPUs are enabling generative AI,” Su mentioned.
The MI300X can use as much as 192GB of reminiscence, which suggests it may match even larger AI fashions than different chips. Nvidia’s rival H100 solely helps 120GB of reminiscence, for instance.
Giant language fashions for generative AI purposes use numerous reminiscence as a result of they run an rising variety of calculations. AMD demoed the MI300x operating a 40 billion parameter mannequin known as Falcon. OpenAI’s GPT-3 mannequin has 175 billion parameters.
“Mannequin sizes are getting a lot bigger, and also you really need a number of GPUs to run the newest massive language fashions,” Su mentioned, noting that with the added reminiscence on AMD chips builders would not want as many GPUs.
AMD additionally mentioned it could provide an Infinity Structure that mixes eight of its M1300X accelerators in a single system. Nvidia and Google have developed related programs that mix eight or extra GPUs in a single field for AI purposes.
One purpose why AI builders have traditionally most popular Nvidia chips is that it has a well-developed software program package deal known as CUDA that allows them to entry the chip’s core {hardware} options.
AMD mentioned on Tuesday that it has its personal software program for its AI chips that it calls ROCm.
“Now whereas it is a journey, we have made actually nice progress in constructing a robust software program stack that works with the open ecosystem of fashions, libraries, frameworks and instruments,” AMD president Victor Peng mentioned.
