Microsoft makes OpenAI’s latest model available on Windows

Microsoft is now allowing Windows users to run OpenAI's new open model on their devices.

Microsoft offers OpenAI’s latest model for Windows users

Microsoft has unveiled the gpt-oss-20b model, which is now available through the Windows AI Foundry for local running on compatible PCs. This model is lightweight and is designed for tasks such as code execution and tool use, making it suitable for creating autonomous assistants. Users will need a device with at least 16GB of VRAM, typically provided by high-end Nvidia or Radeon GPUs. The model, although currently exclusive to Windows, is expected to launch on macOS in the future. This rollout coincides with similar efforts from Amazon to incorporate OpenAI's models into its cloud offerings. This is notable as it marks the first instance of running OpenAI models directly on Windows, intensifying competition in the AI infrastructure space between Microsoft and Amazon.

The introduction of Microsoft’s gpt-oss-20b model represents a significant shift for both the company and the broader AI landscape. By enabling local execution of this lightweight model, Microsoft not only enhances its Windows offerings but also positions itself against key competitors like Amazon, which is also adapting to OpenAI’s advancements. The insistence on using high-end hardware reflects both a push for quality performance and a potential barrier to entry for casual users. This evolution can influence how developers integrate AI into applications and workflows, pushing the boundaries of local control over AI-powered tools.

Highlights

This development may signal further evolution in how AI tools are utilized in everyday workflows.