OpenAI has launched GPT-5.6, a new model family aimed at enterprise work, coding, cybersecurity, and daily productivity tools.
The new lineup includes three variants: Sol, Terra, and Luna. Sol is positioned as the strongest option, Terra sits in the middle, and Luna is the lower-cost model. OpenAI is also tying the release to Microsoft 365 Copilot, where GPT-5.6 is described as the preferred model for productivity apps.
What Is New
- GPT-5.6 is available across ChatGPT, Codex, and the OpenAI API.
- The model family includes Sol, Terra, and Luna for different cost and performance needs.
- OpenAI is positioning GPT-5.6 as stronger for coding and cybersecurity work.
- Microsoft 365 Copilot will use GPT-5.6 as a preferred model across workplace apps.
- OpenAI also introduced ChatGPT Work as a workplace-focused AI companion.
The cybersecurity angle is important because advanced models are increasingly being tested for defensive work such as threat modeling, code review, patching, and blue-team exercises. That makes capability improvements useful, but it also raises questions about misuse if similar skills are applied offensively.
Why The Copilot Update Matters
Microsoft has been building more of its own AI model stack, but the GPT-5.6 Copilot update shows OpenAI still has a central role inside Microsoft’s workplace AI products. The practical question is not whether one model powers every task. It is how Microsoft routes work between OpenAI models and its own systems based on cost, speed, and task complexity.
For users, the update matters because AI assistants are becoming embedded inside documents, spreadsheets, presentations, code tools, and desktop workflows. The model race is moving from chatbot demos into the software people already use every day.
Will workplace AI be judged by the best model, or by the assistant that fits most naturally into daily work?