> ## 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.

# Zotero

[Zotero](https://www.zotero.org) is an open-source reference manager. Zotero itself has no built-in AI, but the community plugin [Zotero GPT (Awesome GPT)](https://github.com/MuiseDestiny/zotero-gpt) adds an in-app assistant that connects to the ScitiX Model Inference API through an OpenAI-compatible endpoint. Keep the plugin updated for compatibility with your Zotero version.

The steps below use Zotero GPT 3.1.x on Zotero 7+/9. Newer versions may differ slightly. For available models, such as `glm-5.2` and `DeepSeek-V4-Flash`, and 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 first.

### 1. Install the Zotero GPT plugin

Download the latest `zotero-gpt.xpi` from the [releases page](https://github.com/MuiseDestiny/zotero-gpt/releases).

In Zotero, open **Tools → Plugins**, click the gear icon, choose **Install Plugin From File…**, and select the `.xpi`. Enable it if prompted. A **GPT** icon appears in the toolbar.

### 2. Configure ScitiX

Open the GPT box by clicking the **GPT** toolbar icon. Then type these commands one at a time, pressing Enter after each:

```text theme={null}
/api https://api.scitix.ai/model-api
/secretKey <Your API Key>
/model glm-5.2
```

| Command      | Value                                                                                                               |
| ------------ | ------------------------------------------------------------------------------------------------------------------- |
| `/api`       | ScitiX endpoint without `/v1`: `https://api.scitix.ai/model-api`. The plugin appends `/v1/chat/completions` itself. |
| `/secretKey` | Your API Key created on the [API Keys](https://console.scitix.ai/model-inference/api_keys) page.                    |
| `/model`     | `glm-5.2`                                                                                                           |

<Warning>
  `/api` must not include `/v1`. Zotero GPT appends `/v1/chat/completions` itself, so adding `/v1` would duplicate the path.
</Warning>

Choose a model with native tool calling if you use the plugin's agent or PDF-RAG features.

The active model is shown next to the **Ask** selector in the GPT box, such as `glm-5.2`.

<img src="https://mintcdn.com/scitix-adfc65c2/_I6H7WpAVXPiLIwE/model-inference/media/zotero-chat-reply.png?fit=max&auto=format&n=_I6H7WpAVXPiLIwE&q=85&s=915aa5062ce1f3a8320cedd77ca27497" alt="Zotero GPT — chatting with a ScitiX model (glm-5.2)" width="1000" height="600" data-path="model-inference/media/zotero-chat-reply.png" />

## Verify

In the GPT box, type a question and press Enter. The reply streams from the ScitiX model.

<Info>
  The assistant may identify itself as "Awesome GPT" or a generic model rather than the exact model ID. That is the plugin's own system prompt, not a routing issue; the model shown next to **Ask** is the ScitiX model in use.
</Info>