OpenAI‘s GPT-5.6 model family is now available in Kiro, an AI-native software development agent focused on engineering rigor and quality at scale. The update brings OpenAI‘s latest flagship models—Sol, Terra, and Luna—into Kiro‘s workflows for planning, building, reviewing, and testing software. The stated goal is to help developers produce higher-quality code with fewer iterations and better value per token.

The models are positioned as delivering more useful work per token, with stronger performance per dollar and on-demand capability for complex tasks. In Kiro, these capabilities apply to long-running development work grounded in requirements, codebase, and team standards. Kiro converts high-level intent into clear requirements, technical designs, and executable tasks, giving the model structured context about what the team is building, how the system should work, and what the final implementation must accomplish.

Developers using GPT-5.6 in Kiro can turn product ideas into structured implementation plans, complete complex multi-step coding tasks with greater consistency, apply spec-driven development, work with context from the codebase and team standards, review and refine the model’s work at key checkpoints before changes are implemented, and check correctness using property-based testing.

OpenAI and AWS jointly optimized the Kiro environment and OpenAI models. Testing on Terminal-Bench 2.1 found that GPT-5.6 Terra completed successful tasks in Kiro at roughly 82% cost reduction. Kiro‘s spec-driven approach grounds the model in clear requirements, technical designs, and task context from the start, so it reaches working solutions faster with fewer missteps. The companies say they will continue collaborating to improve OpenAI model performance in Kiro and help developers get more value from AI across the software development lifecycle.