Skip to main content
Kilo Code is an open-source AI coding agent for VS Code, JetBrains, and the CLI, combining features from Cline and Roo. Its OpenAI Compatible provider accepts a custom Base URL and auto-fetches the model list, so it can run on the ScitiX Model Inference API. Supported models include glm-5.2 and MiniMaxAI/MiniMax-M2.7. For the full model list and pricing, see the model catalog.

​
Configure

Create an API Key on the API Keys page first.
  1. Click the Kilo Code icon in the sidebar, then open Settings (⚙️) → Providers.
  2. Under API Provider, select OpenAI Compatible.
  3. Fill in the required fields.
  4. Kilo Code fetches the model list from https://api.scitix.ai/model-api/v1/models automatically. Select glm-5.2, or another model, from the searchable Model dropdown. If the fetch fails, type the model ID manually.
  5. Click Save. Configurations are stored in a Configuration Profile, such as default, so you can save the ScitiX setup and switch providers quickly.
Kilo Code — OpenAI Compatible provider with ScitiX Base URL and model

​
Verify

Give Kilo Code a coding task, such as “create two sum code with python”, with the ScitiX model selected. It plans the task and asks to create or edit files. Approve with Save. A normal reply confirms that the connection works. Kilo Code — creating a file from a coding task

​
FAQ

  • Empty model dropdown: the GET /v1/models fetch failed. Check the Base URL with the /v1 suffix and the API Key, or type the model ID manually.
  • Model does not act / no tool calls: choose a model with native tool calling. The provider page shows model capabilities such as Supports images after selection.