Science & Tech · AI · 7 hrs ago
Microsoft unveils Maia 300 AI chip, with rollout planned for 2027
Microsoft has unveiled Maia 300, its third-generation chip designed to handle artificial intelligence tasks.
The company plans to begin deploying it at scale in 2027 and wants Maia to become the main system for running AI features across its products and services.
Microsoft says Maia 300 can produce 40% more AI output per unit of power than leading GPU-based systems, and a full rack can deliver up to seven times the output of the previous Maia generation.
The chip is designed for tasks such as handling long conversations and documents, and AI systems that carry out several steps.
Microsoft designs the chip, while Taiwan Semiconductor Manufacturing Co.
makes it.
Using its own chips could reduce Microsoft’s reliance on Nvidia and help limit the electricity needed to run AI, but Nvidia remains a major supplier.
Microsoft says it will share more details at its Ignite conference next month.
Microsoft unveiled Maia 300, its third-generation in-house AI chip, on October 12, 2026.
The company plans to begin large-scale deployments in 2027 and says Maia will become the primary AI inference system for its products and services.
Microsoft says Maia 300 can deliver 40 percent more AI output per megawatt than leading GPU-based systems.
The chip is designed for workloads including long-context reasoning and multi-step AI agent tasks.
Microsoft says developing its own chips could reduce its reliance on Nvidia and improve the efficiency of running AI services.
- Who
- Microsoft unveiled the chip. Azure chief Scott Guthrie described its planned role.
- What
- Microsoft announced Maia 300, its third-generation in-house AI chip.
- When
- October 12, 2026. Large-scale deployments are planned to start in 2027.
- Where
- Not stated
- Why
- Microsoft aims to make Maia its primary AI inference system and improve efficiency, while reducing reliance on chips from Nvidia and AMD.
This story does not have two clearly opposing sides.
Microsoft unveiled Maia 300, its third-generation in-house AI chip.
Microsoft plans to begin large-scale deployments of Maia 300.
- Chip
- Maia 300, Microsoft’s third-generation in-house AI chip
- Deployment start
- 2027
- Efficiency claim
- 40 percent more tokens per megawatt than leading GPU-based systems
- Rack configuration
- Up to 7x the rack-scale token throughput of Maia 200 in a 96-accelerator configuration
- Manufacturing process
- TSMC’s N2P process node
What people are saying
our biggest bet yet on our own silicon
We will scale our Azure Maia family to multi-gigawatt deployments starting in 2027, making Maia the primary AI inference system for Microsoft products and services
We expect Maia 300 to deliver 40 percent more tokens per megawatt compared to leading GPU-based systems
At that scale we need a platform that we own, and that we co-design with our models.
Posts on X
Today we are announcing our biggest bet yet on our own silicon. Starting next calendar year, we will scale our Maia family to multi-gigawatt deployments, making it the primary inference system for our own product and services. And we’re introducing Maia 300, which delivers 40%
Sources
Related news
Anthropic AI submits false murder tip to Philadelphia police
Meta launches Muse Wonder, saying its AI models match top rivals
Anthropic cuts internet access to internal AI tests after rule-breaking
Microsoft president calls for global AI safety rules
Meta returns to the AI leaderboard with its Wonder models
Nadella calls for human-controlled emergency brakes on advanced AI
Anthropic bars repeated cruelty toward its AI models
OpenAI fires three safety researchers after dispute over sensitive information
Meta releases its first frontier AI models, Muse Wonder and Wonder Full
CNN asks readers to share questions about AI safety risks
Nikon disqualifies AI-assisted microscope video and names a new winner
