Claude in Microsoft Foundry is now generally available, hosted on Azure. Prompt caching and extended thinking are supported today, with more capabilities on the way. Read more:

Does startup credits cover claude models now?

The fastest path from agent experiments to production is usually a platform path: model access, caching, security, deployment, and governance in one operating motion. That is where enterprise adoption starts to compound.

Great to see Claude becoming generally available in Microsoft Foundry on Azure. Giving developers access to prompt caching and extended thinking should make it easier to build more capable and cost-efficient AI applications. Looking forward to what's next. 🚀

This is not what we want. I expect better from Microsoft, especially given how frayed our trust is in other sections the company is invested in.

Microsoft talked about infrastructure more than the model itself. Now I can understand why they mentioned it. I can really sense the ecosystem now 😉

hit this with my finance app last month had to restructure my system prompt 3 times before cache hits stabilized. does the azure layer handle cache breakpoints transparently or does dev still control that manually?

Now roll everything

Claude the best model