GPT-5.6 family lands in AWS Kiro, cutting Terminal-Bench costs by 82%

GPT-5.6 登陆 Kiro，为开发者提升性价比

OpenAI brought the full GPT-5.6 family—Sol, Terra, and Luna—into AWS's coding agent Kiro. Kiro turns high-level intent into specs, designs, and tasks, then lets the model plan, build, review, and test. On Terminal-Bench 2.1, GPT-5.6 Terra hit an ~82% cost reduction while completing tasks successfully. The post doesn't disclose token pricing or latency figures, only 'stronger performance per dollar.' I'd discount that 82% a bit: it's a co-optimized internal benchmark; real-world gains depend on your codebase and workflow fit.

Why it matters: OpenAI brings GPT-5.6 to AWS's Kiro coding agent with a concrete 82% cost reduction on Terminal-Bench 2.1 — substantive. But it's an official blog with no third-party validation, and the audience is limited to AWS developers, so resonance is weak. Score at the low end of featu...