AI coding in terminal
Kilo CLI brings AI-powered coding assistance directly to your terminal. Connect it to QwenCloud via Token Plan (Individual), Token Plan (Team Edition), or pay-as-you-go billing.
In addition to the options above, you can also connect using a configuration file.
Replace
Replace
Replace
To add more models, append them in the same format within
After saving the configuration, restart Kilo CLI, type
If you encounter errors during configuration, refer to the FAQ documentation for your billing plan:
Add the model ID to
The API Key does not match the Base URL, or the key belongs to another billing plan. Confirm that the key type matches
When using qwen4-max-preview, the default output limit is 32000; to allow longer output, set
KILO_EXPERIMENTAL_OUTPUT_TOKEN_MAX. Changing limit.output alone does not raise the limit.When using qwen4-max, the default single-request output limit is 32000. To allow longer output, set
KILO_EXPERIMENTAL_OUTPUT_TOKEN_MAX; changing only limit.output does not raise the limit. qwen4-max has a 1M-token context. The maximum output is 131,072 tokens in non-thinking mode and 262,144 tokens for thinking and the final answer combined. The defaults are reasoning_effort=high and thinking_budget=131072.Quick start
1
Start Kilo CLI
2
Connect the provider
Enter
/connect, select the provider for your billing plan, and enter the API key:- Token Plan (Individual): select Alibaba Token Plan (China) and enter the Token Plan (Individual) API Key.
- Token Plan (Team Edition): select Alibaba Token Plan (China), assign a seat, and enter the Token Plan (Team Edition) API Key.
- Pay-as-you-go: select Alibaba (China) and enter your QwenCloud API Key.
3
Select a model
Enter
/models and select the model you want.4
Verify
Configure with a configuration file
In addition to the options above, you can also connect using a configuration file.
- macOS / Linux:
~/.config/kilo/config.json - Windows:
C:\Users\<username>\.config\kilo\config.json
YOUR_API_KEY with the API key for your plan and add the configuration to config.json.
Free quota and billing:
- First-time users get a free quota. See Free quota for details.
- Enable Free quota only to prevent unexpected charges.
Token Plan (Individual)
Replace YOUR_API_KEY with the Token Plan (Individual) dedicated API Key. For available models, see Token Plan (Individual) supported models.
qwen3.8-max thinking mode:
- thinking: Supports both enabled and disabled (hybrid thinking mode).
- temperature: Defaults to 0.6 in thinking mode. Values below 0.6 are automatically adjusted to 0.6.
- reasoning_effort: Controls reasoning depth. Options: xhigh, medium, low. Default: xhigh.
Token Plan (Team Edition)
Replace YOUR_API_KEY with the Token Plan (Team Edition) dedicated API Key. For available models, see Token Plan (Team Edition) supported models.
qwen3.8-max thinking mode:
- thinking: Supports both enabled and disabled (hybrid thinking mode).
- temperature: Defaults to 0.6 in thinking mode. Values below 0.6 are automatically adjusted to 0.6.
- reasoning_effort: Controls reasoning depth. Options: xhigh, medium, low. Default: xhigh.
Pay-as-you-go
Replace YOUR_API_KEY with your QwenCloud API Key. For available models, see supported models.
The following uses the Singapore (ap-southeast-1) endpoint. For another region, replace ap-southeast-1 and use an API key from that region. See Regions and endpoints.
models.
Verify configuration
After saving the configuration, restart Kilo CLI, type /models, search for "QwenCloud", and select the model you want to use.
For more tips and common commands, see the Kilo Code official documentation.
FAQ
Error codes
If you encounter errors during configuration, refer to the FAQ documentation for your billing plan:
- Token Plan: Token Plan FAQ
- Pay-as-you-go: Error codes and troubleshooting
The model I need is not in the list
Add the model ID to models in config.json in the same format, save the file, and reopen Kilo CLI. For supported models, see the supported model list for your billing plan.
Requests return 401 Unauthorized
The API Key does not match the Base URL, or the key belongs to another billing plan. Confirm that the key type matches baseURL in the configuration, then save the configuration again.