> ## Documentation Index
> Fetch the complete documentation index at: https://docs.scitix.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# LazyLLM

> Learn how to use LazyLLM OnlineChatModule with a custom OpenAI-compatible endpoint and build low-code LLM applications on ScitiX models.

[LazyLLM](https://github.com/LazyAGI/LazyLLM) is an open-source low-code LLM application development framework from SenseTime. Its `OnlineChatModule` supports custom OpenAI-compatible endpoints, so you can connect LazyLLM to ScitiX Model Inference.

This guide uses `glm-5.2` as an example. For the full model list and pricing, see [Models](https://console.scitix.ai/model-inference/models).

## Configure

Create an API Key on the [API Keys](https://console.scitix.ai/model-inference/api_keys) page. After installing `lazyllm`, create an online model with `source="openai"` and point it to ScitiX:

```python theme={null}
import lazyllm

m = lazyllm.OnlineChatModule(
    source="openai",
    model="glm-5.2",
    url="https://api.scitix.ai/model-api/v1",
    api_key="<YOUR_API_KEY>",
)
```

| Field     | Description                                                                                     |
| --------- | ----------------------------------------------------------------------------------------------- |
| `source`  | Use `openai` to call through the OpenAI-compatible protocol.                                    |
| `model`   | ScitiX model ID, such as `glm-5.2`                                                              |
| `url`     | ScitiX OpenAI-compatible base URL with `/v1`: `https://api.scitix.ai/model-api/v1`              |
| `api_key` | Your API Key created on the [API Keys](https://console.scitix.ai/model-inference/api_keys) page |

## Verify

Call the model. A normal response confirms that the connection works. This configuration has been verified with **lazyllm 1.3.0**.

```python theme={null}
print(m("Introduce yourself in one sentence."))
```

## FAQ

* **`source` and `url`**:
  Use `source="openai"` and set `url` to the base URL with `/v1`: `https://api.scitix.ai/model-api/v1`.
* **Model IDs**:
  The `model` value is passed through to the upstream service. If the model ID is incorrect, the upstream service returns `404 model_not_found`. Use IDs from the [Models](https://console.scitix.ai/model-inference/models) page.
* **Tool or agent capabilities**:
  Choose a model that supports tool calling. `glm-5.2` supports tool calling.