A selection of the most important recent news, articles, and papers about AI Chipsets.
General News, Articles, and Analyses
Inside Maia 100: Revolutionizing AI Workloads with Microsoft’s Custom AI Accelerator
Author: Sherry Xu
(Tuesday, August 27, 2024) “At Hot Chips 2024, Microsoft shared specifications on Maia 100, Microsoft’s first-generation custom AI accelerator designed specifically for large-scale AI workloads deployed in Azure. Vertically integrated to optimize performance and reduce costs, the Maia 100 system includes a platform architecture featuring custom server boards with tailor-made racks and a software stack built to increase performance and cost efficiency for advanced AI capabilities on services like Azure OpenAI Services.”
AMD will unify Radeon and Pro graphics to battle Nvidia’s dominance | PCWorld
Author: Mark Hachman
(Monday, September 9, 2024) “A top executive said that it plans to combine its CDNA enterprise graphics architecture with the RDNA consumer GPUs to simplify development and take on Nvidia‘s CUDA.”
The Battle Begins For AI Inference Compute In The Datacenter – The Next Platform
Author: Timothy Prickett Morgan
(Tuesday, September 10, 2024) “The difficulty in breaking into AI training among the cloud builders and hyperscalers is one reason why the very many AI chip startups in the world have not set the world on fire with system sales using their chippery and software stacks, despite the dearth of NvidiaGPUs. But these same startups – Cerebras Systems, SambaNova Systems, and Groq are the important ones – are now thinking they have a real shot at building a hardware business – either through direct system sales or through rentals under a cloudy delivery model – because the pressure is shifting to AI inference.”
SambaNova launches AI inference cloud – DCD
https://www.datacenterdynamics.com/en/news/sambanova-launches-ai-inference-cloud/
Author: Georgia Butler
(Tuesday, September 10, 2024) “Chip and artificial intelligence (AI) model developer SambaNova Systems has launched an AI cloud offering. Dubbed SambaNova Cloud, it provides cloud-based AI inference services using the SN40L AI chip.”
Groq To Deploy AI In Saudi Arabia With Aramco, Aims To Match 2023 Microsoft-OpenAI Capacity By 2025 – Benzinga
Author: Rounak Jain
(Wednesday, September 11, 2024) “Nvidia Corp. (NASDAQ:NVDA) rival Groq announced on Tuesday that it will deploy AI in Saudi Arabia in 2024, in partnership with Aramco Digital, at the ongoing Global AI Summit (GAIN) in Riyadh, Saudi Arabia.”
Intel gets boost from Raimondo, effort to spur U.S. chip production | CNBC
Author: Seema Mody
(Thursday, September 12, 2024) “Intel has been badly beaten down of late, losing 60% of its value this year as it struggles to find its way in the booming artificial intelligence market while aggressively building out fabs in the U.S. The company is turning to Commerce Secretary Gina Raimondo for help.”
SambaNova, Cerebras, and Groq Clash Over Token Speed in Wake of OpenAI o1 Launch
Author: Siddharth Jindal
(Sunday, September 15, 2024) “The battle for token speed is intensifying as SambaNova, Cerebras, and Groq push the limits of inference performance.”
