Read a tweet about being broad in narrow markets. Made me wonder who are the closest alternatives/ substitutes to Nvidias GPUs in the next 3 years?

Chris Nheu

The CEO of Arm, which licenses chip designs to Apple and Nvidia, says he sees room for Nvidia competitors.


https://www.theinformation.com/articles/why-arms-ceo-sees-room-for-nvidia-competitors?offer=rtsu-engagement-25&utm_campaign=RTSU%3A+Why+Arm%E2%80%99s+CEO&utm_content=7295&utm_medium=email&utm_source=cio&utm_term=4983&rc=gzsl71

SoftBank-owned Arm has signed up Meta Platforms as a customer of a new server chip for cloud and artificial intelligence computing, the Financial Times reported. The chip wouldn’t compete directly with AI chips from Nvidia and AMD, but it would compete with chips made by customers like Amazon Web Services.

The Arm server chip will use technology from Ampere Computing, a chip startup led by former Intel President Renee James, according to the report. SoftBank began acquisition talks with Ampere last September and is close to completing the deal, according to other reports.

The chip is a departure for Arm, which licenses server chip designs to companies such as Amazon, Apple, and Qualcomm. It is also risky, as an Arm-developed server chip would put it in competition with those customers.

https://www.ft.com/content/95367b2b-2aa7-4a06-bdd3-0463c9bad008?utm_campaign=%5BREBRAND%5D+%5BTI-AM%5D+Th&utm_content=1095&utm_medium=email&utm_source=cio&utm_term=124

SoftBank-owned Arm has signed up Meta Platforms as a customer of a new server chip for cloud and artificial intelligence computing, the Financial Times reported. The chip wouldn’t compete directly with AI chips from Nvidia and AMD, but it would compete with chips made by customers like Amazon Web Services.  The Arm server chip will use technology from Ampere Computing, a chip startup led by former Intel President Renee James, according to the report. SoftBank began acquisition talks with Ampere last September and is close to completing the deal, according to other reports.  The chip is a departure for Arm, which licenses server chip designs to companies such as Amazon, Apple, and Qualcomm. It is also risky, as an Arm-developed server chip would put it in competition with those customers.  https://www.ft.com/content/95367b2b-2aa7-4a06-bdd3-0463c9bad008?utm_campaign=%5BREBRAND%5D+%5BTI-AM%5D+Th&utm_content=1095&utm_medium=email&utm_source=cio&utm_term=124

SoftBank-owned group secures Meta as one of its first customers for new initiative

Arm to launch its own chip in move that could upend semiconductor industry

Sam Mendoza

The CEO of Arm, which licenses chip designs to Apple and Nvidia, says he sees room for Nvidia competitors.   https://www.theinformation.com/articles/why-arms-ceo-sees-room-for-nvidia-competitors?offer=rtsu-engagement-25&utm_campaign=RTSU%3A+Why+Arm%E2%80%99s+CEO&utm_content=7295&utm_medium=email&utm_source=cio&utm_term=4983&rc=gzsl71

Cloud and chip executives have predicted for two years that Nvidia’s dominance in artificial intelligence server chips would crater, but that hasn’t come to pass. Yet Rene Haas, CEO of Arm Holdings, whose chip designs power part of Nvidia’s systems as well as those of major competitors, feels ...

Why Arm’s CEO Sees Room for Nvidia Competitors

Nate

OpenAI is building its own chips, getting as far as ‘taping’ a design out with TSMC. Given OpenAI’s massive historical reliance on third-party hardware (Nvidia) and compute (Microsoft), building its own silicon could make sense. So long as what OpenAI cooks up is sufficiently better for its needs than what it buys today to warrant the focus cost.

https://www.reuters.com/technology/openai-set-finalize-first-custom-chip-design-this-year-2025-02-10/

OpenAI is building its own chips, getting as far as ‘taping’ a design out with TSMC. Given OpenAI’s massive historical reliance on third-party hardware (Nvidia) and compute (Microsoft), building its own silicon could make sense. So long as what OpenAI cooks up is sufficiently better for its needs than what it buys today to warrant the focus cost.  https://www.reuters.com/technology/openai-set-finalize-first-custom-chip-design-this-year-2025-02-10/

OpenAI is pushing ahead on its plan to reduce its reliance on Nvidia for its chip supply by developing its first generation of in-house artificial-intelligence silicon.

Exclusive: OpenAI set to finalize first custom chip design this year

Tasia

Nvidia takes historic $600B nosedive because of DeepSeek 

https://www.cnbc.com/2025/01/27/nvidia-sheds-almost-600-billion-in-market-cap-biggest-drop-ever.html?utm_source=Iterable&utm_medium=email&utm_campaign=newsletter-20241212

Nvidia takes historic $600B nosedive because of DeepSeek   https://www.cnbc.com/2025/01/27/nvidia-sheds-almost-600-billion-in-market-cap-biggest-drop-ever.html?utm_source=Iterable&utm_medium=email&utm_campaign=newsletter-20241212

Nvidia shares plunged 17% on Monday, resulting in a market cap loss of close to $600 billion, the biggest drop ever for a U.S. company.

Nvidia sheds almost $600 billion in market cap, biggest one-day loss in U.S. history

Nvidia this year continued to outpace its rivals and dominate the market for GPUs, the chips used to develop #Artificial Intelligence . The company’s data center business, which includes sales of its GPUs and networking gear, generated nearly $100 billion in revenue in the 12 months ended in October. That’s up 66% from the same period a year prior.

Investor enthusiasm for Nvidia’s financial results lifted its market capitalization above $3 trillion. Nvidia’s biggest customers include cloud computing providers and AI startups such as Elon Musk’s xAI, who use the chips to train and run large AI models.

Nvidia grew faster than Advanced Micro Devices, its primary rival in GPUs, allowing it to gain market share. AMD’s data center revenue, which includes sales of its AI chips, totaled $11 billion in the 12 months ended in September, up roughly 54% from the same period a year earlier.

Broadcom, which designs AI chips for Google, Apple and OpenAI, may now be Nvidia’s biggest rival. Broadcom reported $12.2 billion in revenue from designing custom AI chips and selling networking gear for AI, up 220% from the same period last year, as its customers look to reduce their reliance on Nvidia.

 #Data is beautiful

Artificial Intelligence

Data is beautiful

Nvidia this year continued to outpace its rivals and dominate the market for GPUs, the chips used to develop #Artificial Intelligence . The company’s data center business, which includes sales of its GPUs and networking gear, generated nearly $100 billion in revenue in the 12 months ended in October. That’s up 66% from the same period a year prior.  Investor enthusiasm for Nvidia’s financial results lifted its market capitalization above $3 trillion. Nvidia’s biggest customers include cloud computing providers and AI startups such as Elon Musk’s xAI, who use the chips to train and run large AI models.  Nvidia grew faster than Advanced Micro Devices, its primary rival in GPUs, allowing it to gain market share. AMD’s data center revenue, which includes sales of its AI chips, totaled $11 billion in the 12 months ended in September, up roughly 54% from the same period a year earlier.  Broadcom, which designs AI chips for Google, Apple and OpenAI, may now be Nvidia’s biggest rival. Broadcom reported $12.2 billion in revenue from designing custom AI chips and selling networking gear for AI, up 220% from the same period last year, as its customers look to reduce their reliance on Nvidia.   #Data is beautiful

Chris Nheu

Read a tweet about being broad in narrow markets. Made me wonder who are the closest alternatives/ substitutes to Nvidias GPUs in the next 3 years?

A group of likeminded people in The Most Important Thing are talking about this.