GPT-OSS: Now solve the problem of reasoning-coding on a laptop for free, OpenAI moves towards open sourc

GPT-OSS: Seeing the increasing demand for open source around the world, OpenAI has also launched its new open source model. With which users will be able to use it for free on laptops. The company has launched two versions in this new model. Which can easily operate from reasoning to web browsing, coding and other tasks.

OpenAI has launched a new open source GPT model GPT-OSS. Seeing the increasing demand for open source model, OpenAI has launched the open-weight language model GPT-OSS after almost 6 years. Actually, due to security reasons, the company was not launching such a model. But due to low cost and customization, the demand for open source is increasing among developers. In such a situation, this is considered to be a big change of the company in the direction of open source.

Especially at a time when China's DeepSeek and other companies are also moving fast in the field of AI. This new model of OpenAI is specially for laptops. Users can download it for free and use it by customizing it according to their needs. Let's know what is its specialty.

Two versions launched

OpenAI has launched the GPT-OSS model in two versions. Which includes GPT-OSS-20B and GPT-OSS-120B. Both models support 131,072 token contexts. Regarding the new model, OpenAI has confirmed that both models use the Mixture of Experts (MoE) architecture. Which allows them to activate only a few parameters per token. Such as about 5.1 billion for the 120b model, ensuring greater efficiency and responsiveness.

GPT-OSS-20B: A compact model with around 21 billion parameters that can run with low latency and on consumer hardware such as laptops with 16GB of RAM. It is also ideal for mobile inference or slim desktops.

GPT-OSS-120B: A larger, more powerful model with around 120 billion parameters that matches the performance of OpenAI's own o4-mini. It is intended for high-level reasoning and production workflows. However, it will require a 24GB+ VRAM GPU. It can be run on NVIDIA GPUs.

These models can do these tasks

OpenAI says that these models can do everything from reasoning to web browsing, coding and operating agents. The company has not yet given any information about the release of future versions of GPT-OSS. But the company has expressed hope that GPT-OSS will be used by small developers as well as companies. Because, they will also have the facility to access powerful AI models.

Why is it important?

The demand for open-source AI models is increasing rapidly. In such a situation, this step of OpenAI is being considered important. Especially when companies like Meta, Alibaba, DeepSeek and Mistral have already entered this race. In such a situation, it was necessary for OpenAI to maintain its developer base.