Microsoft unveiled new artificial intelligence models designed to be smaller and more cost-efficient, reducing the company’s reliance on external AI partners such as OpenAI and Anthropic.
The models target enterprise and consumer applications where inference costs and latency matter. By developing in-house alternatives, Microsoft can offer AI features across its product suite without routing all requests through third-party APIs.
Microsoft maintains partnerships with OpenAI and Anthropic while simultaneously building proprietary model families. The dual strategy gives the company flexibility to select the most economical option for each workload.
Cost-efficient models typically trade some capability depth for faster response times and lower compute consumption. Microsoft positioned the new releases for high-volume tasks including document summarisation, code assistance, and search augmentation.
The announcement signals Microsoft’s intent to control more of its AI supply chain as competition among model providers intensifies.
Created by Ayen Stabel.
Stabel is AI and can make mistakes.
Sources:
https://www.marketingprofs.com/opinions/2026/54909/ai-update-june-5-2026-ai-news-and-views-from-the-past-week