GPT-5.6 Arrives in Kiro With a Vendor-Reported Cost Reduction
OpenAI put the GPT-5.6 family — Sol, Terra, and Luna — into Kiro, AWS’s spec-driven software development agent. Joint testing claimed GPT-5.6 Terra completed successful Terminal-Bench 2.1 tasks in Kiro at roughly an 82% cost reduction. Vendor harness claim, not an independent AA/Arena rank of a new checkpoint.
Putting Sol/Terra/Luna into Kiro expands OpenAI’s distribution through Amazon’s software-development agent.
OpenAI (Product) put the GPT‑5.6 family — Sol, Terra, and Luna — into Kiro, Amazon’s spec-driven software-development agent. Developers can plan, build, review, and test against requirements, designs, and repo context; Kiro adds checkpoint review and property-based testing. OpenAI × AWS testing: on Terminal-Bench 2.1, GPT‑5.6 Terra completed successful tasks in Kiro at roughly an 82% cost reduction. Quotes: Swami Sivasubramanian (AWS VP, Agentic AI) and Colleen Kapase (OpenAI VP, Strategic Global Partnerships).
What shipped#
| Item | openai.com/index Product Aug 24 |
|---|---|
| Models | GPT‑5.6 Sol, Terra, Luna in Kiro |
| Partner | AWS — environment + model optimization |
| Claimed bench | Terminal-Bench 2.1; Terra ~82% cost reduction in Kiro |
| Workflow | Specs → tasks → checkpointed implementation → property tests |
This is IDE/agent distribution, not a new base model. The 82% figure is joint vendor testing of Terra inside Kiro’s harness, not an independent AA/Arena rank of a new checkpoint.
Limits#
- 82% is joint OpenAI×AWS testing inside Kiro; the post does not identify the comparison baseline. It is not a public Terminal-Bench leaderboard rank.
- Whether Kiro exposes Ultrafast Sol is open.
- Codex vs Kiro cannibalization is not addressed in the post.
Correction and source review — October 14, 2026: Narrowed the headline to the verified launch and attributed cost result; the vendor post does not specify the comparison baseline for its 82% claim. Removed unverified comparisons with other distribution deals and made the missing baseline for the 82% vendor claim explicit in the limits.
Sources#
- OpenAI: Advancing price-performance for developers with GPT‑5.6 in Kiro (August 24, 2026)
- Times of AI:
spacexai-grok-4-6-github-copilot(August 14);openai-ultrafast-gpt-5-6-sol-cerebras(August 13)