Kimi Code CLI Configuration Guide
¶ 1. Obtain an API Key
Open the 费思量-API application list, enter an available application, and copy the API Key.
¶ 2. Install Kimi Code
macOS / Linux:
curl -fsSL https://code.kimi.com/kimi-code/install.sh | bash
Windows PowerShell:
irm https://code.kimi.com/kimi-code/install.ps1 | iex
You can also install it via npm:
npm install -g @moonshot-ai/kimi-code
¶ 3. Configure 费思量-API
Edit ~/.kimi/config.toml and add the following configuration:
default_model = "acedatacloud/MODEL_ID"
[providers.acedatacloud]
type = "openai"
base_url = "https://api.fesilent.com/v1"
api_key = "your API Key"
[models."acedatacloud/MODEL_ID"]
provider = "acedatacloud"
model = "MODEL_ID"
Go to the model catalog to view model names (for example, claude-opus-5, gpt-6-sol), and replace all MODEL_ID values with the same name; for precise validation, use the id value in the /v1/models response. This value is the model name used by the API model parameter, not the UUID of the service or API. type = "openai" uses the OpenAI Chat Completions protocol.
The configuration file contains the Key; on macOS / Linux, it is recommended to restrict permissions:
chmod 600 ~/.kimi/config.toml
¶ 4. Verify and Start
kimi doctor
kimi provider list
kimi -p "Reply only OK"
Run kimi after entering the project to start using it. If it returns 401, check the API Key; if it indicates model not found, confirm that default_model, the model table name, and the model field use the same ID.