OpenAI's Gpt-OSS EXPLAINED For Users: How To Use GPT-OSS 120B And 20B On Microsoft's Windows?

ChatGPT's parent, OpenAI, has launched two state-of-the-art open-weight language models namely GPT-OSS-120b and GPT-OSS-20B. Following which, Bill Gates-backed Microsoft introduced the GPU optimized gpt-oss-20B model variants to Windows devices. How does it impact users?

These models are available under the Apache 2.0 license and are designed to be used within agentic workflows with exceptional instruction following, tool use like web search or Python code execution, and reasoning capabilities—including the ability to adjust the reasoning effort for tasks that don't require complex reasoning and/or target very low latency final outputs.

What Is GPT-OSS-120b?

This model achieves near-parity with OpenAI o4-mini on core reasoning benchmarks, while running efficiently on a single 80 GB GPU.

What Is GPT-OSS-20B?

While the gpt-oss-20b model delivers similar results to OpenAI o3‑mini on common benchmarks and can run on edge devices with just 16 GB of memory, making it ideal for on-device use cases, local inference, or rapid iteration without costly infrastructure.

As per the official statement, both models also perform strongly on tool use, few-shot function calling, CoT reasoning (as seen in results on the Tau-Bench agentic evaluation suite) and HealthBench (even outperforming proprietary models like OpenAI o1 and GPT‑4o).

In its statement, OpenAI said, "These models are compatible with our Responses API⁠(opens in a new window) and are designed to be used within agentic workflows with exceptional instruction following, tool use like web search or Python code execution, and reasoning capabilities-including the ability to adjust the reasoning effort for tasks that don't require complex reasoning and/or target very low latency final outputs. They are entirely customizable, provide full chain-of-thought (CoT), and support Structured Outputs⁠(opens in a new window)."⁠

It added, "Safety is foundational to our approach to releasing all our models, and is of particular importance for open models. In addition to running the models through comprehensive safety training and evaluations, we also introduced an additional layer of evaluation by testing an adversarially fine-tuned version of gpt-oss-120b under our Preparedness Framework⁠(opens in a new window). gpt-oss models perform comparably to our frontier models on internal safety benchmarks, offering developers the same safety standards as our recent proprietary models."

GPT-OSS On Windows:

Following the release of gpt-oss models, Microsoft announced that it is thrilled to bring GPU optimized gpt-oss-20B model variants to Windows devices.

Microsoft said, "This milestone brings powerful, open-source reasoning models to Windows developers, with support for local inference. You can try it out in Foundry Local or AI Toolkit for VS Code (AITK) and start using it in your applications today."

How To Use GPT-OSS-20b On Windows?

Get gpt-oss-20B up and running on your Windows device in just a few minutes using Foundry Local or AI Toolkit!

To get started with Foundry Local:

Step 1: Install Foundry Local via WinGet (recommended) using the following command: winget install Microsoft.FoundryLocal

Note: As an alternative, Foundry Local can also be installed from GitHub.

Step 2: Open your Terminal and run the model from the Foundry Local CLI with the following command: foundry model run gpt-oss-20B

Step 3: Start sending Foundry Local your prompts!

To get started with AI Toolkit for VS Code:

Step 1: If you don't have it already, install Visual Studio Code.

Step 2: Install AI Toolkit extension.

Step 3: Open Model Catalog and download the model gpt-oss-20B.

Step 4: Open the Model Playground, load the model, and start sending it prompts!

After exploration via either tool, you can modify prompts, tune inference parameters, and integrate into your app using the Foundry Local SDK.