Microsoft Unveils Maia 300 AI Chip in Major Push to Reduce Reliance on Nvidia
24 Sources
[1]
Microsoft's new AI chip makes it less reliant on Nvidia's GPUs
Microsoft is announcing its third-generation Maia in-house AI chip today, designed to be the "primary AI inference system" for Microsoft's products and services. Maia 300 not only improves the cost and efficiency of running AI workloads in the cloud, it's also more capable of handling the latest
[2]
Microsoft plans multi-gigawatt rollout of next-gen Azure Maia AI accelerator racks
Microsoft has lifted the lid on the next generation of its Azure Maia custom AI chips, saying it will usher in a new phase of its compute infrastructure where everything is closely integrated to meet the demands of agentic inference. The Redmond giant says its Azure Maia 200, announced earlier
[3]
Microsoft plans huge deployment of its own AI chips, with an eye on saving costs
SAN FRANCISCO, Oct 12 (Reuters) - Microsoft (MSFT.O), opens new tab announced the next generation of its custom AI chip on Monday and said it plans to deploy potentially more than a million of them next year to back its internal AI services. The Maia 300, as the chip is called, is the third
[4]
Microsoft to rely more on homegrown AI chips, reducing dependence on Nvidia
* Microsoft said it will eventually have multiple gigawatts worth of its Maia AI chips. * The company intends to make Maia its primary architecture for running AI models. * The next-generation Maia 300 chip enters production in 2027. In this article * MSFT Follow your favorite stocksCREATE
[5]
Microsoft makes 'biggest bet yet' on its own silicon, unveils new AI chip to reduce reliance on Nvidia
Microsoft plans to run most of its AI services on chips it designs itself, a shift that could save the company billions of dollars and reduce its dependence on industry juggernaut Nvidia. The company on Monday unveiled Maia 300, the third generation of its custom AI chip, saying it will begin
[6]
Microsoft says Maia 300 chip will run its AI services from 2027
Maia will become the main system for running AI models in Copilot and other Microsoft products, the company said. Microsoft will make its own Maia chips the primary system for running AI in its products starting in 2027. The company set out the plan on Monday as it unveiled the Maia 300, the third
[7]
Microsoft Maia 300 96-chip rack hits 40% more tokens per megawatt
Microsoft plans to break free from NVIDIA and deploy its next-generation Maia 300 AI accelerator in 2027, targeting 40% more tokens per megawatt than leading GPU-based systems. The chip will power Microsoft's expanding AI infrastructure, with a 96-accelerator rack configuration designed to deliver
[8]
Microsoft's Maia 300 swaps raw compute for a 2.5x memory bandwidth gain as it moves to make in-house silicon its main AI engine while claiming a 40% efficiency bump -- could going the Google route be the pin that pops the Nvidia AI bubble?
* Microsoft will make Maia its main inference engine at multi-gigawatt scale, following Google and Amazon into in-house silicon * Microsoft's upcoming Maia 300 trades a modest 36% compute gain for 2.5x Maia 200's memory bandwidth, betting on memory-bound decode, with production in 2027 *
[9]
Every major Nvidia customer is now a competitor as Microsoft reveals new Maia chip
Why it matters: Microsoft's plan to power its AI products with custom silicon shows how every major Nvidia customer is now a competitor, seeking lower costs, improved performance and more control. * The shift is particularly significant for inference -- the computing needed to generate AI
[10]
Microsoft Maia 300 AI chip to lead inference workloads
Starting with the third-generation Maia 300, Microsoft says it will rely on its own silicon to handle incoming AI model requests Microsoft $MSFT unveiled its third-generation Maia 300 AI accelerator on Monday and said it will use its homegrown chips as the primary architecture for running AI
[11]
Microsoft unveils Maia 300, its most powerful AI chip yet - and it might make using Teams, Copilot and Cowork much better at last
* Microsoft reveals Maia 300, its latest AI chip * Designed for "inference-first", Maia 300 offers huge improvements in efficiency and power * Maia 300 will begin deploying in 2027 Microsoft has revealed Maia 300, the next generation of its Azure hardware as it prepares to scale up its in-house
[12]
Microsoft debuts multichiplet Maia 300 inference accelerator
Microsoft debuts multichiplet Maia 300 inference accelerator Microsoft Corp. today pulled back the curtains on Maia 300, its newest internally developed artificial intelligence accelerator. The chip is optimized for inference. Microsoft estimates that it can process 40% more tokens per megawatt
[13]
Microsoft unveils Maia 300 AI chip
Microsoft unveiled its third-generation Maia 300 AI chip on Monday and said the Maia family will become its primary inference system for products and services starting in 2027. The company plans to scale Maia deployments to gigawatt levels. Reuters reported that the rollout could involve more than
[14]
Microsoft prepares to put gigawatts of its chips in data centers
(Bloomberg) -- Microsoft is gearing up to install the latest version of its own chips in data centers, with the aim of powering its AI software using mostly homegrown hardware. The company expects to stock multiple gigawatts of its Maia chips in its server farms, with the newest iteration, the
[15]
Microsoft bets big on new Maia 300 AI chip to reduce reliance on NVIDIA GPUs
Microsoft has unveiled Maia 300, the third generation of its in-house AI chip, and wants it to become the main engine behind its AI products and services. Large-scale deployment starts in 2027, with Reuters reporting that it could involve more than a million chips. CEO Satya Nadella has described
[16]
Microsoft Introduces Maia 300 AI Chip With These Improvements
Microsoft recently began wider deployment of the Maia 200 AI chip Microsoft introduced its first AI inference chip in 2023. Later, in January 2026, the tech giant introduced the Maia 200 AI chip, which is capable of running OpenAI's AI models and Microsoft's MAI model. The inference chip is
[17]
Microsoft Azure is not taking cost-cutting measures, says software giant spokesperson amidst AI overhaul reports
Microsoft has announced the next generation of its custom AI chip and said it plans to deploy potentially more than a million of them next year to back its internal AI services. Microsoft Azure is not going for any lay off like move. Microsoft spokesperson said that Azure leadership has not
[18]
Microsoft plans huge deployment of its own AI chips, with an eye on saving costs
The Maia 300, as the chip is called, is the third generation of the company's custom chips designed to handle AI data crunching such as coding and carrying out complex business tasks on a user's behalf. Its predecessor is already running Microsoft's in-house AI models, as well as models from OpenAI
[19]
Microsoft Plans Multi-Gigawatt Azure Maia Expansion to Meet Growing AI Demand
The multigenerational Azure Maia program has become a foundational element of Microsoft's AI infrastructure strategy. As we move into the era of Super Intelligence, Microsoft is entering a new phase of infrastructure. We will scale our Azure Maia family to multi-gigawatt deployments starting in
[20]
Microsoft Plans to Deploy 1 Million Maia 300 AI Chips in 2027
Microsoft has unveiled Maia 300, its third-generation AI chip, and plans to install more than one million units in 2027. The company wants to run its AI services faster and at a lower cost. The chip was announced on Monday, October 12. Maia 300 is built for heavy AI work, such as writing code and
[21]
Microsoft to scale custom AI chips in multi-gigawatt push to cut costs By Investing.com
Investing.com -- Microsoft Corporation (NASDAQ:MSFT) plans to deploy custom-designed artificial intelligence processors across its global datacenter footprint at a multi-gigawatt scale starting in 2027, marking a aggressive push to curb reliance on third-party hardware vendors and lower operating
[22]
Microsoft plans to rely on its own Maia chips for AI infrastructure
On its official blog, Microsoft has announced plans to use its own Azure Maia chip family for multigigawatt deployments as part of its AI infrastructure starting in 2027, with the aim of making it the primary inference system for its products and services. The Redmond, Washington-based group noted
[23]
Microsoft plans huge deployment of its own AI chips, with an eye on saving costs
SAN FRANCISCO, Oct 12 (Reuters) - Microsoft announced the next generation of its custom AI chip on Monday and said it plans to deploy potentially more than a million of them next year to back its internal AI services. The Maia 300, as the chip is called, is the third generation of the company's
[24]
Microsoft unveils Maia 300 AI chip to reduce reliance on Nvidia, promises 40 pct greater efficiency
HIGHLIGHTS Microsoft unveils Maia 300 AI chip for better efficiency. The chip promises 40 per cent more AI tokens per watt. Maia 300 deployments are expected to begin in 2027. Microsoft has officially unveiled Maia 300, its third-generation in-house artificial intelligence chip, designed to run
Share
Copy Link
Microsoft announced its third-generation custom AI chip, Maia 300, designed to become the primary AI inference system across its products. Starting production in 2027, the chip delivers 40% more tokens per megawatt than leading GPU-based systems and seven times the rack-scale throughput of its predecessor. This marks Microsoft's biggest bet yet on self-designed silicon as it aims to reduce dependence on Nvidia.
Microsoft Bets Big on Self-Designed Silicon with Maia 300
Microsoft announced its third-generation custom AI chip, Maia 300, on Monday, marking what CEO Satya Nadella called "our biggest bet yet on our own silicon."
5
The Azure Maia AI accelerator is designed to become the "primary AI inference system for Microsoft products and services," according to Scott Guthrie, executive vice president of Microsoft's cloud group.1
Production begins in 2027, with multi-gigawatt deployments planned to reduce reliance on Nvidia and AMD chips that currently power most of Microsoft's AI infrastructure.The move represents a fundamental shift in how Microsoft approaches AI infrastructure. While the company has spent years depending on Nvidia's graphics processing units, it's now positioning Microsoft's homegrown AI chips as the foundation for running AI inference workloads at scale. This vertical integration in AI infrastructure mirrors strategies already deployed by competitors like Google with its Tensor Processing Units and Amazon with its Inferentia and Trainium chips.
4
Massive Efficiency Gains Drive Strategic Pivot
Maia 300 delivers significant performance improvements that translate directly into operational cost savings. The chip produces 40% more tokens per megawatt compared to leading GPU-based systems, according to Microsoft.
3
In a full 96-accelerator rack-scale system configuration, it delivers up to seven times the rack-scale token throughput of Maia 200.1
Source: GeekWire
These efficiency improvements are critical as power supply has become one of the biggest constraints on AI expansion. Each Maia 300 rack containing 96 chips uses about 167 kilowatts of power, compared with 142 kilowatts used by Nvidia's GB300 AI server racks with 72 GPUs.
3
"It's basically delivering more intelligence for every watt, every dollar, and every byte of memory by optimizing across the entire system," said Rani Borkar, president of Azure hardware systems and infrastructure at Microsoft.3
Advanced Chiplet Architecture Powers Next-Gen Performance
Microsoft employed a sophisticated chiplet architecture to achieve its efficiency gains. Built on TSMC's N2P process node, the Maia 300 design separates compute, memory controller, and I/O functions into specialized components.
1
The central computing processor uses 2-nanometer technology from Taiwan Semiconductor Manufacturing Co for key computing functions, while other chiplets use cheaper 3-nanometer tech for functions like moving data.3
Source: Axios
Each Maia 300 features eight HBM4 memory stacks providing up to 18 TB/s of peak memory bandwidth, optimized for executing large models and long-context workloads.
2
This approach differs from Nvidia's design, which uses one massive computing chip surrounded by high-bandwidth memory chips.3
Designed for Emerging AI Workloads and Agentic Tasks
Maia 300 is specifically engineered for the new types of AI workloads emerging today: long-context reasoning, latency-sensitive conversational experiences, and sustained, multi-step agentic tasks.
1
Microsoft describes it as "a complete intelligent system" with an inference-first architecture optimized for long-context, memory-intensive and decode-heavy AI workloads.2
The system-level approach integrates software, silicon, physical rack infrastructure, networking, and datacenters co-designed to operate as a single platform. Microsoft insists this integration is necessary for agentic workloads, which involve multi-step execution, long-context decoding and tool use, fundamentally changing "the computational profile and arithmetic intensity of inference."
2
Maia 200 Already Powering Production Workloads
The current-generation Maia 200, announced in January, is already running production AI inference workloads across Microsoft's ecosystem. It powers Microsoft Copilot, GitHub Copilot, Microsoft Foundry, and Microsoft's own AI models.
1
Guthrie stated that "Maia 200 is the best performing AI infrastructure in our fleet, measured in real world metrics for AI services."1
The chip is being deployed across the United States, Europe and Australia, with GitHub Copilot and Microsoft Foundry set to move onto it by the end of the year.
5
Nadella told analysts in July that Maia 200 "delivers 30% better performance per dollar than the latest generation hardware in our fleet" and is handling OpenAI models.4
Related Stories
Strategic Implications for Microsoft's AI Ambitions
Maia 300 plays a critical role in Microsoft's broader AI model development strategy. "We have to work across the full stack, all the way down to the silicon," said Microsoft AI CEO Mustafa Suleyman. "At that scale we need a platform that we own, and that we co-design with our models. That's why running our Microsoft AI models on top of the Maia chip is a critical focus for our team."
1
This vertical integration could help Microsoft improve performance and operational costs while catching up with AI leaders like Google DeepMind, OpenAI, and Anthropic. Finance chief Amy Hood noted at a Goldman Sachs conference that use of its own chips has helped with "margin improvement over the course of a decade."
4
Massive Scale Deployment Plans and Market Impact
Microsoft plans potentially more than a million Maia chips deployed next year to back its internal AI services.
3
The company expects to spend more than $175 billion in capital expenditures this calendar year.3
Evercore ISI analysts estimate that Maia will represent about 10% of Microsoft's spending on AI chips by the first quarter of 2027.4
Source: The Register
However, the transition will be gradual. Alistair Speirs, a general manager at Microsoft, declined to provide a specific timeline, noting "we reserve a little bit of optionality there, because this is an industry that's changing very fast."
4
Microsoft will share more details at its Microsoft Ignite conference next month in San Francisco.2
Nvidia Partnership Continues Despite Strategic Shift
Despite the push toward self-designed silicon, Microsoft and Nvidia remain significant partners. Nadella and Nvidia CEO Jensen Huang shared the stage last week to launch Windows PCs built on Nvidia chips.
5
Nvidia remains a key supplier for Microsoft's data centers, particularly for training models rather than inference.4
Nadella said on the company's July earnings call that Microsoft will be among the first cloud providers to deploy next-generation systems based on Nvidia's Vera Rubin and AMD's Helios.
5
When it comes to training models, Microsoft partners with both Nvidia and AMD, according to Speirs.4
This balanced approach allows Microsoft to maintain flexibility while building greater control over its Azure cloud infrastructure.References
Summarized by
Navi
Related Stories
Recent Highlights
1
OpenAI's AI Models Disprove 27-Year Exponential Time Hypothesis, Rendering Decades of Research Obsolete
Science and Research
2
Google DeepMind unveils Genie 4 world model with audio and 2-minute interactive simulations
Technology
3
Google Unveils Gemini 4 Carbon AI Model With Major Cybersecurity Gains, Beating Claude Opus 5.5
Technology
Recent Highlights
Today's Top Stories
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Follow topics that matter to you and stay ahead.
