product updateMoonshot AI

Kimi Code CLI v2.7.0 adds native Amazon Bedrock provider with standard AWS credentials

TL;DR

Kimi Code CLI v2.7.0, Moonshot AI's open source coding agent, adds a native bedrock provider. It resolves the standard AWS credential chain, including named profiles and SSO, so users no longer need a separate Amazon Bedrock API key. AWS demonstrates the setup with Kimi K3.1 Pro Highspeed.

3 min read
0

Moonshot AI's Kimi Code CLI v2.7.0 adds a native bedrock provider, letting the open source coding agent authenticate to Amazon Bedrock with the standard AWS credential chain instead of a separately managed Amazon Bedrock API key. AWS detailed the setup in a post on its Machine Learning blog.

What changed

Older versions of Kimi Code CLI reached Amazon Bedrock through a generic provider with type = "openai" or type = "openai_responses". These use the OpenAI-compatible Chat Completions or Responses APIs. That setup required a regional base_url and an Amazon Bedrock API key, short-term or long-term, as an extra credential to obtain and protect.

The new provider changes three things:

  • AWS authentication workflow: it resolves temporary credentials through the standard AWS CLI credential chain. Users can rely on the default chain or select a named profile, including one backed by SSO.
  • Inferred API format and endpoint: the provider infers the API format and regional endpoint from the model ID. Both can be overridden per model.
  • Explicit keys still win: a key supplied through api_key or api_key_env takes precedence over the credential chain.

How it works

According to AWS, the provider handles authentication and routing in four steps:

  1. Resolve AWS credentials from the standard chain, unless an Amazon Bedrock API key is configured. AWS_PROFILE selects a named profile.
  2. Generate a short-lived bearer token using @aws/bedrock-token-generator, renewed when necessary. No AWS access key or manually generated Bedrock key needs to go into the Kimi Code config.toml.
  3. Build the regional endpoint from AWS_REGION or AWS_DEFAULT_REGION. For Kimi models, this resolves to the recommended OpenAI-compatible Chat Completions path.
  4. Call the selected inference profile. The example uses the global inference profile global.moonshotai.kimi-k3.1. The global. prefix selects the profile, and AWS_REGION determines which regional endpoint receives the request.

The ephemeral API keys are valid for up to 12 hours or the AWS session length, whichever is shorter.

Example model and requirements

AWS uses Kimi K3.1 Pro Highspeed as the walkthrough model. It supports streaming, configurable reasoning, image input, and client-side tool calling. AWS says global cross-Region inference gives the broadest availability and lowest cost. Pricing is listed on the Amazon Bedrock pricing page.

Requirements:

  • An AWS account with access to the target model and inference profile in a supported source Region
  • Kimi Code CLI v2.7.0 or later (check with kimi --version)
  • The AWS CLI; the optional aws login workflow requires v2.32.0 or later
  • An IAM user or assumed role permitted to invoke the model and inference profile

AWS provides an example IAM policy scoped to the Kimi K3.1 Pro Highspeed model and its global inference profile. It includes bedrock:CallWithBearerToken restricted to SHORT_TERM bearer tokens, plus bedrock:GetInferenceProfile, bedrock:InvokeModel, and bedrock:InvokeModelWithResponseStream. The foundation-model ARN without a Region or account ID is intentional and required for global cross-Region routing. The project/default statement is needed only with the openai_responses protocol, not the default openai Chat Completions protocol.

For SSO through AWS IAM Identity Center, users run aws configure sso --profile YOUR_PROFILE_NAME, then aws sso login --profile YOUR_PROFILE_NAME.

AWS recommends short-term credentials over long-term ones and advises a dedicated IAM role for Kimi Code CLI, separate from regular AWS activities, under least-privilege principles.

What this means

This is a packaging and authentication change, not a new model. For teams already managing AWS access through profiles or IAM Identity Center, Kimi Code CLI now fits that workflow directly, removing the need to issue and store a separate Bedrock API key. AWS notes the change reduces reliance on manually managed keys but does not change the user's responsibility to protect AWS credentials and scope permissions. Teams that need a particular API format or endpoint can still override the inferred defaults per model.

Related Articles

product update

Amazon Polly voice cloning is generally available as a gated release in seven languages

Amazon Polly voice cloning is now generally available as a gated release. It builds a custom voice on the Polly generative engine from a few minutes of one consenting speaker's audio, in seven languages. Building a voice is free, and usage is billed pay-as-you-go.

product update

Nous Research brings open-source Hermes Agent to Microsoft Teams, adds Collective Wisdom skill sharing

Nous Research says companies will soon be able to run its open-source Hermes Agent directly in Microsoft Teams, with agents sandboxed in Azure Container Apps. A new feature, Collective Wisdom, lets employees share reviewed skills with colleagues, and the shared knowledge stays inside the company.

product update

BioMap launches AI drug safety system with xTrimoTox, reports 0.9637 AUROC on liver injury

BioMap launched an AI-assisted nonclinical drug safety assessment system on November 19. It combines virtual-cell models, toxicity prediction and evaluation services, and the company reports AUROC of 0.9637 for drug-induced liver injury on a selected benchmark task.

product update

OpenAI tests ChatGPT dots with a small group of Plus users, rolls out to all Business Standard

OpenAI has started testing ChatGPT dots with a small number of ChatGPT Plus subscribers and is rolling it out to all ChatGPT Business Standard users. Dots launched in September for Pro, Business Premium and Enterprise plans. Usage allowances for the new tiers have not yet been clarified.

Comments

Loading...