OpenAI released the free gpt-oss-20b model for PCs, which Microsoft has made available on Windows AI Foundry, with macOS support to follow. Users will need PCs or laptops with a minimum of 16GB of VRAM, compatible with top Nvidia or Radeon GPUs. This model is designed for code execution and tool usage, well-suited for building autonomous assistants and integrating AI into real-world applications, even in limited bandwidth scenarios. Microsoft optimized the model for local inference, hinting at future support for more devices.
OpenAI's new free and open GPT model, gpt-oss-20b, designed for PCs, is now made accessible to Windows users via Windows AI Foundry.
A PC or laptop must have at least 16GB of VRAM, necessitating high-performance Nvidia or Radeon GPUs for optimal usage of the gpt-oss-20b model.
The gpt-oss-20b model is optimized for code execution and tool use, making it suitable for developing autonomous assistants and integrating AI in workflows.
Microsoft's integration of OpenAI's model with its local inference capabilities indicates potential future expansions to enhance support for various devices.
Collection
[
|
...
]