Skip to content
context Tools and apps

Kimi Code CLI gains native support for Amazon Bedrock with AWS login

clearly official source

Moonshot AI has released version 2.7.0 of the Kimi Code CLI tool with a native Bedrock provider, which enables login via the standard AWS credential chain (profiles, SSO) instead of a separate API key for Amazon Bedrock.

Moonshot AI, the creator of the open-source Kimi models, has released version 2.7.0 of the open-source Kimi Code CLI tool. The new feature is a native "bedrock" provider for Amazon Bedrock, which replaces the previous solution using a generic provider of type openai or openai_responses that required a separate Amazon Bedrock API key.

The native provider can obtain credentials directly from the standard AWS CLI credential chain, including named profiles and single sign-on (SSO). Instead of a long-lived API key, it thus uses short-term (ephemeral) credentials valid for a maximum of 12 hours or for the duration of the AWS session, whichever occurs first. According to AWS, this reduces the need for manually created and managed API keys, but responsibility for protecting AWS credentials and setting appropriate permissions remains with the user.

As an example model, the article uses Kimi K3.1 Pro Highspeed, which, according to the source, supports streaming, configurable reasoning, image input, and client-side tool calling. The login workflow via AWS CLI requires AWS CLI version 2.32.0 or newer. For production deployment, AWS recommends a dedicated IAM role for Kimi Code CLI, separate from regular AWS activity, with permissions set according to the principle of least privilege.

The source article further describes specific configuration steps (IAM policies, editing the config.toml file, a test request), but these are not complete in the available text. Details can be found in the source article.

What changed

Why it matters

For teams that already use AWS and want to work with Kimi models via Amazon Bedrock, the need to create and protect a separate long-lived API key is eliminated. Login can be handled the same way as for other AWS services (profiles, SSO, IAM roles), which simplifies operations and reduces the risk of leakage or misuse of a permanent key, since the new credentials are short-term.

Two audiences, two different impacts

What this means

01

For individuals

Developers using Kimi Code CLI with Amazon Bedrock can switch to logging in via a standard AWS profile or SSO and avoid manually managing a separate Bedrock API key.

What to do When using Kimi Code CLI with Amazon Bedrock, set up login via an AWS CLI profile or SSO instead of a separate API key.
More practical updates →
02

For a business

Companies running Kimi Code CLI on AWS can unify access control through existing IAM policies and reduce the risk associated with long-lived API keys thanks to short-term credentials.

Risks and compliance
What to decide Check whether the deployment of Kimi Code CLI on AWS uses a dedicated IAM role with permissions set according to the principle of least privilege.
More business impacts →
Amazon Bedrock AWS credentials coding agent Kimi Code CLI LLM integration Moonshot AI

Check the original

Event sources

clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.

1
AWS Machine Learning Blog primary source · first detected Use Kimi Code CLI with models on Amazon Bedrock