• Faster and more capable of handling long, complex coding tasks

• Can create apps, games, and websites ready to use with greater ease

OpenAI has launched a new programming model called GPT-5.3-Codex, describing it as their most advanced tool yet to help developers complete tasks on computers.

The model combines advanced coding abilities with deeper understanding of professional tasks and runs about 25% faster than the previous version.

It can carry out long, complex projects that require research, tool use, and multiple steps, while allowing users to interact with it in real time without losing context.

OpenAI noted that early versions of the model even helped improve its own development by detecting errors, managing tests, and streamlining deployment.

The company confirmed that the model’s capabilities go beyond writing and reviewing code, extending to most tasks developers and professionals perform on a computer.

It has shown strong results on specialized tests measuring coding skills and automated work in real-world environments.

In practical use, the model can build complex apps and games from scratch in just a few days, with improvements in design and usability.

It also better understands requests for simple websites, automatically producing pages with enhanced features and sensible defaults, giving users a stronger starting point to turn their ideas into real projects.