> ## 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.

# LibreChat

> Learn how to connect LibreChat to ScitiX Model Inference so a self-hosted ChatGPT-style web UI can use ScitiX models.

[LibreChat](https://github.com/danny-avila/LibreChat) is an open-source, self-hosted ChatGPT-style UI supporting many providers. It reaches the ScitiX Model Inference API through its **custom endpoints** feature, configured in `librechat.yaml`.

Default model here is `glm-5.2`; see the [Models](https://console.scitix.ai/model-inference/models) page for the full list and pricing.

Deploy LibreChat per the [official guide](https://www.librechat.ai/docs).

## Configure

Providers in LibreChat are defined server-side in `librechat.yaml`, not in the UI. Create an API key on the ScitiX [API Keys](https://console.scitix.ai/model-inference/api_keys) page, then add ScitiX as a custom endpoint:

```yaml theme={null}
version: 1.2.1
cache: true
endpoints:
  custom:
    - name: "ScitiX"
      apiKey: "${SCITIX_API_KEY}"
      baseURL: "https://api.scitix.ai/model-api/v1"
      models:
        default:
          - "glm-5.2"
        fetch: false
      titleConvo: true
      titleModel: "glm-5.2"
      modelDisplayLabel: "ScitiX"
```

| Field            | Value                                                                                                                 |
| ---------------- | --------------------------------------------------------------------------------------------------------------------- |
| `name`           | Label shown in the endpoint selector, e.g. `ScitiX`                                                                   |
| `apiKey`         | `${SCITIX_API_KEY}` — resolved from the environment at request time                                                   |
| `baseURL`        | ScitiX OpenAI-compatible endpoint: `https://api.scitix.ai/model-api/v1`                                               |
| `models.default` | ScitiX model IDs to offer, e.g. `glm-5.2`. Pick a model with native tool calling for LibreChat's agent/tool features. |
| `models.fetch`   | `false` — ScitiX does not expose OpenAI's `/models` list format, so list models here instead of fetching              |

Pass the key in the environment LibreChat reads (its `.env`, or the container env). Make sure the config file is picked up (`CONFIG_PATH=/path/to/librechat.yaml`, or the default `./librechat.yaml`):

```bash theme={null}
SCITIX_API_KEY=<Your API Key>
```

## Verify

Restart LibreChat, sign in, and open a new chat. In the top-left **endpoint selector** choose **ScitiX** (model `glm-5.2`), send a message — a normal reply confirms the connection.

<img src="https://mintcdn.com/scitix-adfc65c2/AblxxMM8pLmVYmdt/model-inference/media/librechat-chat-reply.png?fit=max&auto=format&n=AblxxMM8pLmVYmdt&q=85&s=a22e61a7df7f7e1cf0f67a7f0d94e7d6" alt="LibreChat replying via the ScitiX endpoint on glm-5.2" width="2982" height="1430" data-path="model-inference/media/librechat-chat-reply.png" />

LibreChat has no UI to add a custom provider, so the ScitiX endpoint must be defined in the config file, not the settings screen.

Verified on LibreChat v0.8.7.