Skip to main content

Kilo Code

Connect Kilo Code in VS Code to AI Studio's OpenAI-compatible API.

This quickstart guide shows you how to connect Kilo Code in VS Code to Hyperstack AI Studio using the OpenAI-compatible API. You will configure Kilo Code to use a model hosted in AI Studio, authenticate with your API key, and validate the integration with a test prompt.

Full Guide

For the complete walkthrough including detailed testing workflows and advanced configuration, see How to Integrate Hyperstack AI Studio with Kilo Code in VS Code.

Why Integrate Kilo Code with Hyperstack AI Studio​

Kilo Code is a multi-agent development assistant built directly into VS Code. It goes beyond autocomplete by understanding your repository context and developer intent. Instead of a single chat interface, it provides specialized agents such as Ask, Code, Debug, and Architect, each focused on different parts of the software development lifecycle.

By connecting Kilo Code to Hyperstack AI Studio, you control the model powering these agents. Rather than relying on a fixed third-party backend, you can choose any base or fine-tuned model for your workflow. This is especially valuable for teams that require consistent behavior, private deployments, or custom training on internal codebases.

With this integration, you can:

  • Use any base or fine-tuned model hosted in AI Studio
  • Maintain full control over your model backend
  • Standardize model usage across your team
  • Leverage OpenAI-compatible APIs without additional adapters

Hyperstack AI Studio provides a unified platform for training, fine-tuning, deploying, and monitoring large language models. It exposes an OpenAI-compatible API, allowing Kilo Code to connect using the built-in "OpenAI Compatible" provider option without a proxy layer.

In this setup, Hyperstack AI Studio handles inference, scaling, and authentication, while Kilo Code serves as the developer interface inside VS Code, orchestrating agent workflows and sending structured requests to your selected model.

For advanced workflows and architecture details, refer to the full integration guide.

How to Install and Connect Kilo Code to Hyperstack AI Studio​

Follow these steps to install Kilo Code and connect it to a model hosted in Hyperstack AI Studio.

  1. Install Visual Studio Code​

    If you do not already have VS Code installed:

  2. Install the Kilo Code Extension​

    • Open VS Code
    • Go to Extensions or press Ctrl + Shift + X
    • Search for Kilo Code
    • Click Install
    • Accept required permissions when prompted

    After installation, confirm that Kilo Code appears in the VS Code sidebar and open the Kilo Code panel.

  3. Retrieve API Credentials from Hyperstack AI Studio​

    You will need the Base URL, Model ID, and API key.

    a. Log in to the Hyperstack Console
    b. Navigate to the AI Studio Playground
    c. Select and test a model such as openai/gpt-oss-120b
    d. Open the API tab and copy the Base URL and Model ID
    e. Go to the API Keys page
    f. Click Generate New Key, name it, and copy the key securely

  4. Configure Kilo Code with Hyperstack​

    In VS Code:

    a. Open the Kilo Code panel
    b. Click the settings icon
    c. Select Use your own API key
    d. Choose provider: OpenAI Compatible

    Enter the following values exactly as retrieved from Hyperstack:

    • Base URL: https://console.hyperstack.cloud/ai/api/v1
    • Model Name: openai/gpt-oss-120b or your deployed model ID
    • API Key: Paste your generated Hyperstack API key

    Click Save and complete the setup. Optionally, name the profile for easier identification if you manage multiple providers.

  5. Validate the Integration​

    a. Open the Kilo Code Ask panel
    b. Confirm the active model is openai/gpt-oss-120b or your selected model
    c. Send a test prompt such as:

    Hi

    If you receive a valid response, the integration is successful and Kilo Code is communicating with your Hyperstack-hosted model.

  6. Test Code Generation​

    To confirm agent functionality:

    • Open the Code or Agent panel
    • Submit a prompt such as:
    Create a Python file named hello_world.py that prints "Hello, World!"

    Verify that the file is created in your workspace. This confirms that agent workflows are correctly using your Hyperstack model.

Next Steps

Explore advanced workflows, team configuration, and model optimization in the full guide: How to Integrate Hyperstack AI Studio with Kilo Code in VS Code.


Back to top

On this page