Claude Sonnet 4.6 is now available in Kiro
Nima Kaviani
Product
Starting today, you can use Claude Sonnet 4.6 in the Kiro IDE and CLI. Sonnet 4.6 is a full upgrade from Sonnet 4.5 for software development — approaching Opus 4.6 intelligence while being more token efficient, which means faster, high-quality task completion across your workflows.
The coding model you’ll want to use all day
Sonnet 4.6 is smart enough that you don’t need to hand-hold it through complex codebases. It handles iterative workflows like building features, refactoring, or debugging without losing quality when you’re guiding the model as you go. Sonnet 4.6 delivers a great experience on interactive, back-and-forth development work.
When you’re working through a spec with multiple tasks, Sonnet 4.6 maintains context and precision across long sessions. It fills both lead agent and subagent roles in multi-model pipelines, with adaptive thinking and context compaction that give you precise control over how work gets done.
Built for agentic work
Sonnet 4.6 is purpose-built for the kind of multi-step orchestration that modern AI development demands. It handles tool calling, maintains state across long sessions, and manages complex reasoning chains — all at a price point that makes high-volume use practical.
For teams using Kiro powers and custom subagents, Sonnet 4.6 slots in naturally. It can fill both the orchestrator and worker roles without needing different models for different parts of the pipeline.
Availability
Sonnet 4.6 is available as an experimental model to all Kiro Pro, Pro+, and Power customers, logging in with Google, GitHub, AWS BuilderID and AWS IAM Identity Center. Sonnet 4.6 is available in the AWS US-East-1 (Northern Virginia) and AWS Europe (Frankfurt) regions, and similar to Sonnet 4.5 features a 1.3x credit multiplier, making it very cost effective for development work.
Download Kiro or restart the app or CLI to use the new model.