Claude Opus 5 is now available for selection in Kiro IDE / CLI
This page has been translated by machine translation. View original
Introduction
On July 24, 2026, Anthropic announced Claude Opus 5, and Kiro also added support for Opus 5.
Usage via Amazon Bedrock is covered in a separate article.
The supported plans are Kiro Pro, Pro+, Pro Max, and Power, and the supported regions are us-east-1 and eu-central-1 (with cross-region inference support). Since it is being rolled out gradually, please restart your IDE or CLI if it does not appear in the model selector.
Verification Details
| Item | Value |
|---|---|
| Kiro IDE | Version 1.0.212 (VSCode 1.107.1) |
| Kiro CLI | 2.14.1 |
| OS | macOS (Darwin arm64 25.5.0) |
Model Selection in Kiro IDE
In the model selector, it was displayed as 2.20x credits / 1M context / Experimental preview.
Model Selection in Kiro CLI
Output of the /model command:
Select model: type to search
❯ auto 1.00x credits Models chosen by task for optimal usage and consistent quality
claude-opus-5 2.20x credits Experimental preview of Claude Opus 5 model with 1M context window
claude-sonnet-5 1.30x credits Experimental preview of Claude Sonnet 5 model with 1M context window [active]
claude-opus-4.8 2.20x credits Claude Opus 4.8 model with 1M context window
gpt-5.6-sol 2.40x credits Experimental preview of OpenAI GPT 5.6 Sol with 272k context window
gpt-5.6-terra 1.20x credits Experimental preview of OpenAI GPT 5.6 Terra with 272k context window
gpt-5.6-luna 0.60x credits Experimental preview of OpenAI GPT 5.6 Luna with 272k context window
claude-opus-4.7 2.20x credits Claude Opus 4.7 model with 1M context window
(+11 more)
The credit multiplier is the same 2.20x as Opus 4.8/4.7, so switching will not change your credit consumption.
Summary
Claude Opus 5 can be selected as an Experimental preview in both Kiro IDE and CLI. According to the official blog, it is considered the first model to try when running parallel agents, so please give it a try if you need a high-performance agent.
