Microsoft brings GPT-5 and GPT OSS to AI Toolkit for VS Code
Readers help support Windows Report. We may get a commission if you buy through our links.
Read our disclosure page to find out how can you help Windows Report sustain the editorial team. Read more
Microsoft has rolled out AI Toolkit v0.18.3 for Visual Studio Code. It brings full integration of OpenAI’s GPT-5 family of models and the new GPT OSS open-source models.
Developers can now access the latest GPT-5 lineup directly through GitHub, Azure AI Foundry, or OpenAI, with options tailored for different needs. The flagship GPT-5 model focuses on advanced reasoning for analytics, orchestration, and code generation.
GPT-5-mini offers lower cost and latency for real-time tools, while gpt-5-nano is optimized for speed and concise reasoning. There’s also a preview of gpt-5-chat, designed for natural, multimodal, multi-turn conversations with persistent context, ideal for building agent-driven apps.
The update also marks the debut of OpenAI’s GPT OSS within the toolkit, providing developers with open-source options that can run in the cloud or locally.
The gpt-oss-120b model is available via Azure AI Foundry for large-scale deployments, featuring adjustable reasoning effort and structured output support. For edge or Windows setups, gpt-oss-20b-cuda-gpu offers CUDA-optimized local inference with performance-boosting RTN quantization.
AI Toolkit now makes it easier to compare models side-by-side, generate ready-to-run SDK snippets in Python or JavaScript, and deploy to different environments — whether cloud, local, closed-source, or open-source.
With this release, Microsoft is positioning the AI Toolkit as a flexible hub for experimenting with and integrating both cutting-edge and open-source AI models. It will help streamline development from prototype to production.
