Artificial intelligence news.
Straight from the original announcements.

Kimi Code CLI adds native Amazon Bedrock provider, no separate API key needed

November 20, 2026, 01:33By the AI News Bites Editorial TeamSource: AWS
Kimi Code CLI adds native Amazon Bedrock provider, no separate API key needed
Image: AWS

Moonshot AI has added a native bedrock provider to Kimi Code CLI v2.7.0, letting the open source coding agent reach models on Amazon Bedrock with standard AWS credentials instead of a separate Bedrock API key. AWS describes the setup in a post that uses Kimi K3.1 Pro Highspeed as its example model.

Key points
  • Kimi Code CLI v2.7.0 adds a native bedrock provider, according to AWS.
  • It uses the standard AWS credential chain, including named profiles and SSO, instead of a separate Bedrock API key.
  • The provider generates short-term keys valid for up to 12 hours or the AWS session length, whichever is shorter.
  • AWS's example uses Kimi K3.1 Pro Highspeed via the global inference profile global.moonshotai.kimi-k3.1.
  • AWS recommends least-privilege IAM permissions and a dedicated role for the CLI.

From a separate API key to the AWS credential chain

Kimi Code CLI is an open source coding agent from Moonshot AI, the creator of the Kimi series of open-weight foundation models. According to AWS, connecting it to Amazon Bedrock previously meant setting up a generic provider with type = "openai" or type = "openai_responses", which use the OpenAI-compatible Chat Completions or Responses APIs. That route required the regional base_url and an Amazon Bedrock API key, short-term or long-term, which AWS says was an extra credential for customers to obtain and protect.

With v2.7.0, the native provider can resolve credentials through the standard AWS Command Line Interface credential chain. Users can rely on the default chain or pick a named profile, including one backed by single sign-on (SSO). The provider infers an API format and regional endpoint from the model ID, and both can be overridden per model.

AWS notes that it recommends short-term credentials over long-term ones. It adds that the new path reduces reliance on manually managed API keys, but does not change the user's responsibility to protect AWS credentials and scope permissions appropriately.

How the provider works

AWS lists four steps in how the provider handles authentication and request routing:

  • Resolve AWS credentials. Unless a Bedrock API key is configured, the provider uses the standard credential chain. Setting AWS_PROFILE selects a named profile.
  • Generate a short-lived bearer token. The provider uses @aws/bedrock-token-generator to create a short-term Bedrock API key and renews it when needed. No AWS access key or manually generated Bedrock key has to be placed in config.toml, although an explicit key given through api_key or api_key_env takes precedence.
  • Build the regional endpoint. The provider reads AWS_REGION or AWS_DEFAULT_REGION and builds the matching bedrock-runtime endpoint. For Kimi models this resolves to the recommended OpenAI-compatible Chat Completions path.
  • Call the selected inference profile. The example uses the global inference profile global.moonshotai.kimi-k3.1. The global. prefix selects the profile, while AWS_REGION determines the regional endpoint that receives the request.

The generated keys are valid for up to 12 hours or the AWS session length, whichever is shorter. AWS says authentication, authorization, model access and network connectivity must all succeed for a request to complete.

Prerequisites and permissions

The walkthrough uses Kimi K3.1 Pro Highspeed, which AWS says supports streaming, configurable reasoning, image input and client-side tool calling. It is accessed through global cross-Region inference, which AWS says offers the broadest availability and lowest cost. AWS points readers to Amazon Bedrock pricing before use, and to the model card for current Region and inference-profile support.

Users need an AWS account with access to the model, Kimi Code CLI v2.7.0 or later, and the AWS CLI. The optional aws login workflow requires AWS CLI v2.32.0 or later. They also need an IAM identity with permission to invoke the model and inference profile. AWS provides an example policy scoped to the Kimi K3.1 Pro Highspeed model and its global profile only. One statement, covering the default project, is needed only with the openai_responses protocol and not with the default openai Chat Completions protocol.

AWS advises granting permissions by the principle of least privilege, particularly in production, and suggests a dedicated IAM role for Kimi Code CLI, separate from regular AWS activities.

Getting started and what it means

Setup begins with credentials. SSO users run aws configure sso --profile YOUR_PROFILE_NAME and then aws sso login --profile YOUR_PROFILE_NAME. Others can use aws login with AWS CLI v2.32.0 or later, or the default profile. AWS recommends checking the resolved identity with aws sts get-caller-identity. Step 2 is adding the provider and model to ~/.kimi-code/config.toml, where the Region must be set explicitly for Kimi Code even if a profile already defines one.

For teams already standardized on AWS identity tooling, the change means Kimi Code CLI can fit into existing SSO and profile workflows rather than adding a separately managed Bedrock key. Amazon Bedrock offers pay-per-token inference, and AWS says processing stays within the AWS data boundary.

Related news

Airbnb Uses Cost per Task to Tune Coding Agents on Amazon Bedrock
For developersNovember 18, 2026, 22:48

Airbnb Uses Cost per Task to Tune Coding Agents on Amazon Bedrock

Airbnb is using cost per task, alongside model evaluations and production monitoring, to select models and manage coding agents running through Amazon Bedrock. The…

AWS outlines three ways to cut costs in Bedrock agent JSON workloads
For developersNovember 17, 2026, 23:44

AWS outlines three ways to cut costs in Bedrock agent JSON workloads

AWS has outlined three strategies for reducing token consumption, latency and memory overhead in Amazon Bedrock AgentCore agents that process large JSON payloads. In…

Moonshot AI’s Kimi K3.1 Pro Highspeed arrives on Amazon Bedrock
ModelsNovember 11, 2026, 02:07

Moonshot AI’s Kimi K3.1 Pro Highspeed arrives on Amazon Bedrock

Moonshot AI’s Kimi K3.1 Pro Highspeed is now available on Amazon Bedrock, giving developers access to a post-trained Kimi K3 model with a 1-million-token context window…