Enhanced delegation with subagents
For developers leveraging GPT-5.6 within Codex, the model’s ability to delegate work to subagents has been significantly improved. Previous models also had access to this feature, but GPT-5.6 specifically “They are trained to decide for themselves when delegation is helpful.” In the card game demonstration, the model leveraged subagents to divide up the work and complete it in parallel, handling tasks such as art and sound design simultaneously. This autonomous delegation capability is a powerful feature of agent AI, allowing models to manage and optimize complex projects by distributing tasks efficiently.
API model to meet diverse needs
OpenAI also makes the GPT-5.6 series available via its API, offering a variety of models to suit different development needs. Developers can choose from “Flagship model for ambitious agent operations” be “A more balanced model” and “A fast and affordable model suitable for everyday tasks.” This tiered offering allows developers to choose the best model for their specific use case, balancing performance, cost, and complexity. The announcement of GPT-5.6 marks a significant step forward in making more intelligent and autonomous AI available to a broader developer community, enabling the creation of increasingly sophisticated applications.
