---
title: "Initial Setup"
slug: "initial-setup-2"
tags: ["Content Services"]
updated: 2026-10-12T11:43:20Z
published: 2026-10-12T11:43:20Z
canonical: "kb.ctera.com/initial-setup-2"
---

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

# Initial Setup

Before using CTERA Content Services, you have to set up the following:

1. An LLM
2. A collector
3. System settings
4. Sources that you can query
5. Datasets for the sources

After setting up the above you can use these components to create experts to enable querying the sources

## Adding LLMs

A Large Language Model (LLM) is a type of AI deep learning model trained on massive datasets to understand, generate, and summarize text or other content. Based on transformer architectures, LLMs act as next-word predictors.

**To add an LLM:**

1. Sign in as an administrator, using the following address `https://<ip>/admin/login`  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-QO1H7C6A.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)  

The administration dashboard is displayed.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-7YZKYXAZ.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
2. Select **LLMs** in the navigation pane.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-9H5NXOJI.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
3. Click **New LLM**.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-4KS0W4TU.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
4. Under **LLM Details**, enter the following:  

**Name** – A name for the LLM.  

**Provider** – Select an LLM provider from the list.  

**Model** – Select a model from the list.
5. Under **Endpoint**, enter the **API Key**.
6. Under **LLM Status**, the status to **Enabled**.
7. Click **Save Changes**.

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-4Q7FU85F.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)

## Adding a Collector

A collector is designed to gather and organize the massive amounts of data required to train or operate AI models (like LLMs), which require high-quality datasets to learn.

**To add a Colllector:**

1. Select **Collectors** in the navigation pane.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-8E3D24N2.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
2. Click **New Collector**.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-ASMPBK4U.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
3. Enter the following details in the **General** tab:  

**Name** – A name to identify the collector.  

**URI** – The URI to access the collector.  

**Default Collector** – Slide on to make this collector the default collector.  

**Collector Status** – Select whether you want the status to be `Enabled` or `Disabled`
4. Click **Connect** in the URI field to connect to the collector and retrieve file type mappings.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-ULR0U9VV.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
5. For each file type, select the **Reader** if you don't want the default.
6. Click **Save Changes**.

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-L7MQORHM.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)

## Managing System Settings

1. Select **Settings** in the navigation pane.  

The **System** tab for **Settings** is displayed.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-O2VJRVOI.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
2. Enter the following details:  

**Ingest Workers** – The number of ingestor workers for scaling.  

**Embedding Mode** – The embedding mode to use from the drop-down list. The  port must be open. See [Inbound Ports](https://kb.ctera.com/docs/technical-specification).  

**Embedding Model** – The embedding model to use from the drop-down list.  

**Embedding Service URL** – The URL for the embedding mode.  

**Sparse Embedding Model** – The sparse embedding model to use from the drop-down list.  

**FastEmbedService URL** – The fast embed service URL. The  port must be open. See [Inbound Ports](https://kb.ctera.com/docs/technical-specification).  

**Default Model** – The LLM that will be used as the default, from the LLMs defined to CTERA Content Services.  

**Classification Model** – The LLM model to use for data classification. If enabled, select the model from the drop-down list of LLMs defined to CTERA Content Services.  

**Summarize Model** – The LLM to use to create summaries. .  

**Vision-Language Model (VLM)** – The model to use for the VLM. If enabled, select the model from the drop-down list of LLMs defined to CTERA Content Services.  

**Personalized Chat Memory** – Save chat prompts across the user conversations.

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-5DV2Q4LA.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)

## Adding Sources

**To add a source:**

1. Select **Sources** in the navigation pane.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-ZBL3GWFW.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
2. Click **New Source**.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-8HYUCHBX.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
3. Enter the source details:  

**Name** – A name to identify the source.  

**Type** – A type of source from the drop-down list.  

**Host** – The URL to access the data source.  

If the **Type** is `Web` you also specify the number of directories to crawl or 0 for an unlimited number of directories.
4. Enter the authentication details. The required details depend on the **Type**.
5. Enter whether you want the source status to be `Enabled` or `Disabled` and click **Save Changes**.

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-SNDXC97H.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)

## Adding Datasets

**To add a dataset:**

1. Select **Datasets** in the navigation pane.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-JEBJJ4ZB.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
2. Click **New Dataset**.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-WLL7XGDO.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
3. Enter the dataset details:  

**Name** – A name to identify the dataset.  

**Source** – A source from the drop-down list. The source list includes the sources defined in **Sources**.  

**Capabilities** – The search and tagging capabilities to apply to this dataset.  

**Owner** – The dataset owner.  

**Collaborators** – The users who can collaborate on the dataset.  

**Collector** – The type of collector to use.
4. Add content to the dataset. The selected content section is dependent on the source. For example, when the source is a CTERA Portal, the content includes cloud drive folders that belong to a specific owner or without any owner.
  1. Click ![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-9HJ7BSYJ.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D) to add a folder.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-XVNBO12L.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
  2. Click **Users** to expand the users and then select a user and from the list of cloud folders, select a folder.  

The folder content is displayed.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-KU8N86G8.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
  3. Click **Add Folder**
5. Enter the schema details and then the indexing actions required.
6. Click **Save Changes**.

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-JN6U61AK.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)

## Adding an Expert

Experts are task-focused AI agents, that act as *virtual employees* to respond to specific types of enquires using your organization’s curated content. For example, a marketing expert might have access to all the company marketing material and competitive information, while a developer expert might have access to all the CTERA SDK and its examples.

**To create an expert:**

1. Sign in as an administrator, using the following address `https://<ip>/admin/login`  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-QO1H7C6A.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)  

The administration dashboard is displayed.
2. Click **Experts** in the navigation pane.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-J4DQOQN7.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
3. Click **New Expert**.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-GC2SW9BV.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
4. Enter the expert details:  

**Expert Name** – A name to identify the expert.  

**Expert Subtitle** – A short description of the expert.
5. Upload a jpeg or png file to use as the expert avatar.
6. Select the expert owner from the drop-down list and collaborators for this expert, from the drop-down list.  

If you specify any collaborators, only these users will have access to the expert.
7. Slide **Set as Public Expert** on if you want this expert to be available publicly.
8. Slide **Set as Default Expert** on if you want this expert to be the default expert for all users.
9. Under **Expert Settings** specify the **Instructions** for the expert.
10. Select the LLM to use for this expert in the **LLM Model** field.
11. Select the datasets to use for this expert in the **Datasets** field.
12. Slide the tools on under **Expert Tools** that will be used with this expert, for example **Semantic Search** and **Image Generator**.
13. Optionally , select an MCP Server from the **MCP Servers** list.
14. Slide **Expert Enabled** on to make the expert active and available to users.
15. Click **Save**

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-G52TDCJQ.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)

**To test the expert:**

1. In a separate browser tab, enter `https://<ip>/` where  is the IP address of CTERA Data Intelligence to open the end user view.  

The end user dashboard is displayed.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-BHRTIHMZ.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
2. Write a message and verify you receive an answer.

Or:

1. Click the expert.  

![Image](https://cdn.document360.io/30792387-cdcb-4791-aa14-94929bbcd3fe/Images/Documentation/image-Z0T1QBZX.png?sv=2026-02-06&spr=https&st=2026-11-11T04%3A17%3A27Z&se=2026-11-11T05%3A17%3A27Z&sr=c&sp=r&sig=uNDVkEcYDzJ%2F5ScfwQJIAGilx803AMqiRX3vhpGC%2F1E%3D)
2. In the **Expert Settings** panel, click **Test Expert**.

**Note**

In both test methods, if **Set as Public Expert**is not enabled, you have to sign in to CTERA Content Services.