# IDEA Kilo Code

> Connect to the LMU AI API with the Kilo Code plugin in IntelliJ IDEA — Claude uses the Anthropic protocol, GPT uses the Responses protocol.

URL: https://docs.lmuai.ai/docs/tools/kilo-code-idea







## Install the plugin [#install-the-plugin]

1. Open IntelliJ IDEA and go to **Settings → Plugins → Marketplace**
2. Search for **Kilo Code**, and click Install once you find the plugin
3. Restart IDEA after installation finishes

***

## Configure the plugin [#configure-the-plugin]

Open the Kilo Code settings panel and pick the option that matches the type of model you want to use:

<Tabs items="['Chinese models / Claude (Anthropic)', 'OpenAI models (OpenAI Compatible)']">
  <Tab value="Chinese models / Claude (Anthropic)">
    Works for all Claude and Chinese models, such as `claude-opus-5`, `claude-sonnet-5`, `qwen3.8-max-preview`, `deepseek-v4-pro`, `glm-5.2`, and `kimi-k3`.

    | Field                   | Value                                                          |
    | ----------------------- | -------------------------------------------------------------- |
    | **API Provider**        | `Anthropic`                                                    |
    | **Anthropic API Key**   | the `sk-` key generated in the LMU AI console                  |
    | **Use custom base URL** | ✅ checked                                                      |
    | **Base URL**            | `https://api.lmuai.ai`                                         |
    | **Model**               | the latest models such as `claude-opus-5`, `qwen3.7-max`, etc. |

        <img alt="Kilo Code with the Anthropic provider selected, Use custom base URL checked and set to https://api.lmuai.ai, and the model set to claude-opus-4-6" src="__img0" />

    <Callout type="info">
      You must check **Use custom base URL**, otherwise the plugin will request the official Anthropic endpoint instead of the LMU AI relay.
    </Callout>
  </Tab>

  <Tab value="OpenAI models (OpenAI Compatible)">
    Works for all OpenAI models, such as `gpt-5.6-sol`, `gpt-5.6-terra`, `gpt-5.5`, and `gpt-5.4`.

    | Field            | Value                                         |
    | ---------------- | --------------------------------------------- |
    | **API Provider** | `OpenAI Compatible (Responses)`               |
    | **Base URL**     | `https://api.lmuai.ai/v1`                     |
    | **API Key**      | the `sk-` key generated in the LMU AI console |
    | **Model**        | e.g. `gpt-5.5`                                |

        <img alt="Kilo Code with the OpenAI Compatible (Responses) provider selected, Base URL set to https://api.lmuai.ai/v1, and the model set to gpt-5.5" src="__img1" />

    <Callout type="warn">
      The Base URL for OpenAI models &#x2A;*must end in `/v1`**, unlike the Anthropic protocol (no `/v1`) — don't mix the two up.
    </Callout>
  </Tab>
</Tabs>

***

## Model name reference [#model-name-reference]

Not sure which model to enter? Go to the [Model Gallery](/docs/guide/models) to see every available model ID and copy it with one click.

***

## Notes [#notes]

* Using **Claude / Chinese models** → choose the `Anthropic` protocol, with Base URL `https://api.lmuai.ai` (no `/v1`)
* Using **OpenAI models** → choose the `OpenAI Compatible (Responses)` protocol, with Base URL `https://api.lmuai.ai/v1` (with `/v1`)
* For the API Key, enter the `sk-` key generated in the LMU AI console under either protocol
* Once configured, click **Save** in the top-right corner to save the settings and start using it
