Connect Kilo Code

Kilo Code is a coding agent for VS Code. For the terminal version, see Kilo CLI.

Before you start: create a Fortytwo API key and buy credits — see Quickstart, step 1. Give each agent its own key with a Spend limit.

Setting	Value
Base URL	`https://api.fortytwo.network/v1`
Model ID	`qwen/qwen3.8-27b`

Install Kilo Code

In VS Code, open Extensions, search for Kilo Code and press Install on Kilo Code: AI Coding Agent by the verified publisher Kilo Code. The extension is also on the VS Code Marketplace and Open VSX .

VS Code Extensions view: Kilo Code by the verified publisher Kilo Code, Install button

If VS Code asks Do you trust the publisher “Kilo Code”?, press Trust Publisher & Install.

VS Code: Trust Publisher & Install dialog for Kilo Code

Kilo onboarding: choose how you work

Click the Kilo icon in the side bar. On Welcome to Kilo, choose how you want to work: Review first or High autonomy. Either option works with Fortytwo.

Kilo Code: Welcome to Kilo with the Review first and High autonomy options

Add Fortytwo as a provider

Open Settings (the gear at the top of the Kilo panel).
Then the Providers tab → Custom provider → Connect.
Fill in the dialog:

Field	Enter
Provider ID	`fortytwo`
Display name	`Fortytwo`
Provider API	OpenAI Compatible
Base URL	`https://api.fortytwo.network/v1`
API key	Your Fortytwo API key
Models	Press Fetch models and add `qwen/qwen3.8-27b`. Turn on Reasoning and Image

Press Submit.

Kilo Code: Settings gear in the Kilo panel opens Kilo Settings on the Providers tab

Kilo Code: Custom provider row with the Connect button

Kilo Code: Custom provider dialog filled in for Fortytwo, model qwen/qwen3.8-27b with Reasoning and Image on

Set the context size

Kilo needs the model’s context and output limits to manage long sessions. In Providers, open the Fortytwo provider again. In Edit provider, press Edit advanced settings in the JSON config file: kilo.jsonc opens in the editor. Add the limit line to the model:

~/.config/kilo/kilo.jsonc

"qwen/qwen3.8-27b": {
  "name": "qwen/qwen3.8-27b",
  "limit": { "context": 262144, "output": 16384 },
  "reasoning": true
}

Save the file.

Kilo Code: Edit provider dialog with the Edit advanced settings in the JSON config file link

kilo.jsonc: limit line added to the qwen/qwen3.8-27b model of the fortytwo provider

Select the model and start

For a wider chat, press Open in Tab at the top of the editor: Kilo opens in an editor tab.

Click the model selector under the message box. It shows Auto Free until you change it.
Search for Fortytwo.
Choose qwen/qwen3.8-27b and press Select.

The selector now shows qwen/qwen3.8-27b. Type a message or a task and press Enter. The reply shows the model’s Reasoning above the answer.

Kilo Code: message ready to send with qwen/qwen3.8-27b selected

Kilo Code: reply from qwen/qwen3.8-27b with its reasoning

Kilo Code and Kilo CLI share these settings. More options: Kilo OpenAI-compatible provider .

Something not working with your agent, or using one we have not listed? Tell us in the platform: Feedback for ideas and small bugs, Contact support if it blocks your work, or Report a security issue.

Agents

Apps

Terminal

Advanced