Skip to main content

Documentation Index

Fetch the complete documentation index at: https://agent37.com/docs/llms.txt

Use this file to discover all available pages before exploring further.

If you want to run OpenClaw with Kimi instead of OpenAI or Claude, this guide walks you through the full setup inside Agent 37. You will create a Moonshot AI API key, open your hosted OpenClaw terminal, and configure Kimi K2.5 as the active model for your instance. This page is based on the video walkthrough below and turns it into a written setup guide you can follow at your own pace.

​
Video walkthrough

Before you begin:

​
Why use Kimi K2.5 with OpenClaw

Kimi K2.5 is a useful option when you want another strong model provider for coding, long context tasks, or multilingual workflows. If you already run OpenClaw in Agent 37, switching to Kimi is mostly a configuration change inside the terminal.

​
Step 1: Create your Moonshot AI API key

  1. Sign in to the Kimi API Platform.
  2. Open the console or API key management area.
  3. Create a new API key for your workspace.
  4. Copy the key and store it somewhere safe.
Keep this key private. Anyone with the key can consume usage on your Moonshot AI account.

​
Step 2: Open your Agent 37 instance

  1. Go to the Agent 37 dashboard.
  2. Open an existing OpenClaw instance, or create a new one.
  3. Click Terminal.
You will do the full model setup from the hosted Agent 37 terminal. You do not need to configure this on your local machine.

​
Step 3: Start the OpenClaw configuration flow

In Terminal, run: 
openclaw configure
When prompted, choose:
  1. Where will the gateway run? → Local
  2. Select Section to Configure? → Model
  3. Model / Auth Provider → Moonshot AI (Kimi)

​
Step 4: Add your Moonshot AI API key

After selecting the provider:
  1. Choose the Moonshot AI API key authentication option
  2. Paste the API key you created in Step 1
  3. Continue to the model picker
OpenClaw will save the credential to your instance configuration.

​
Step 5: Select a Kimi model

Choose the Kimi model you want OpenClaw to use as the active default model for this instance. A common choice is: 
kimi-k2.5
Moonshot’s Kimi API docs also describe Kimi K2.5 as an OpenAI-compatible API surface, which makes it a practical provider option for OpenClaw workflows.

​
Step 6: Verify the model is active

Run: 
openclaw models list
You should see the selected Kimi model in the output. If you want to switch models later, you can rerun openclaw configure or use the exact model identifier shown by OpenClaw in the model management commands.

​
What to do next

Change default model

Switch between configured model providers as your workflow changes.

OpenAI API

Compare the Kimi setup flow with the OpenAI provider flow.