Microsoft makes Cobalt 200 Arm-based cloud chips generally available
The Azure virtual machines are aimed at cloud, database and AI agent workloads, and are priced above the earlier Cobalt 100 generation.
By Standard Topic. AI-generated from the reporting listed under Coverage. Published . Updated .
This story is no longer being updated. It reflects reporting up to .
What happened
Microsoft has made its Cobalt 200 virtual machines generally available on Azure, powered by its second-generation Arm-based server chip. The company announced the launch at its Ignite conference in San Francisco on Tuesday, Nov 17. The 132-core chip is aimed at cloud-native, database, developer and AI agent workloads, and the VMs are available in 14 regions.
Why it matters
Microsoft claims up to 50% better performance than the previous generation, but the 4-vCPU Linux size lists at $0.196 per hour in East US, 40% above Cobalt 100, so customers need a per-vCPU speedup above 1.4x to cut cost per task.
Timeline
- Nov 17: Microsoft announced at its Ignite conference in San Francisco that Cobalt 200 virtual machines are generally available.
- Nov 17: The VMs became generally available in 14 regions, with the 4-vCPU Linux size listed at $0.196 per hour in East US.
Coverage
- Azure Cobalt 200 VMs Launch: List Price Runs 40% Above Cobalt 100, So Speedup Must Top 1.4x (techtimes.com)
- Microsoft Launches Azure VMs With In-House 'Cobalt 200' CPU, Targeting AI Agents With Up to 1.9x Performance Gains (BigGo Finance)
- Microsoft’s ARM-based Cobalt 200 chips go GA. Why it matters for AI agents. (thenewstack.io)
- Microsoft Azure Cobalt 200 VMs reach general availability, bringing Arm-powered performance to cloud and agentic AI workloads (newsroom.arm.com)
- Azure Cobalt 200 VMs now available (azure.microsoft.com)
More from Technology & AI
- China puts humanoid robots to work serving food and sorting parcels
- Newsom proposes sweeping AI safety rules for Washington
- Google releases Gemini 4 Neon AI model with million-token context
- Microsoft and Intel widen their cloud chip partnership
- Google DeepMind rolls out Genie 4 to select subscribers
- AI helped AMD shorten chip design work, OpenAI says