> ## 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.

# NextChat

[NextChat](https://github.com/ChatGPTNextWeb/NextChat) is a lightweight, cross-platform ChatGPT-style web/desktop client. It overrides the OpenAI base URL through environment variables, so you can point it to the ScitiX Model Inference API and use ScitiX models.

Supported models include `glm-5.2` and `DeepSeek-V4-Flash`. For model pricing, see the [model catalog](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. NextChat reads its OpenAI connection from environment variables. Set these before starting:

```bash theme={null}
export OPENAI_API_KEY="<Your API Key>"
export BASE_URL="https://api.scitix.ai/model-api"
export CUSTOM_MODELS="-all,+glm-5.2,+DeepSeek-V4-Flash"
```

| Variable         | Description                                                                                                                  |
| ---------------- | ---------------------------------------------------------------------------------------------------------------------------- |
| `OPENAI_API_KEY` | Your API Key created on the [API Keys](https://console.scitix.ai/model-inference/api_keys) page.                             |
| `BASE_URL`       | ScitiX endpoint without the `/v1` suffix: `https://api.scitix.ai/model-api`. NextChat appends `/v1/chat/completions` itself. |
| `CUSTOM_MODELS`  | `-all` hides the default models; `+glm-5.2,+DeepSeek-V4-Flash` adds the ScitiX models to the picker.                         |

<Warning>
  `BASE_URL` must not include `/v1`, or NextChat will duplicate the path when it appends `/v1/chat/completions`.
</Warning>

Choose a model with native tool calling if you use plugin or tool features.

The same three variables apply when deploying on Vercel. Set them in the project's environment variables. For Docker, pass them with `-e OPENAI_API_KEY=... -e BASE_URL=... -e CUSTOM_MODELS=...`.

## Verify

Open NextChat, pick a ScitiX model from the selector, and send a message. A normal reply confirms that the connection works.

<img src="https://mintcdn.com/scitix-adfc65c2/_I6H7WpAVXPiLIwE/model-inference/media/nextchat-chat-reply.png?fit=max&auto=format&n=_I6H7WpAVXPiLIwE&q=85&s=e847977427e5c81c99a47473e9e34238" alt="NextChat — chatting with a ScitiX model" width="2880" height="1800" data-path="model-inference/media/nextchat-chat-reply.png" />

## FAQ

* **Model picker empty or defaults showing**: check `CUSTOM_MODELS`. `-all` clears the built-in list, then `+<model>` adds each ScitiX model.