> For the complete documentation index, see [llms.txt](https://docs.cosmocloud.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.cosmocloud.io/resources/vector-search/create-a-vector-search-index.md).

# Create a Vector Search Index

* Head over to the vector search indexes option in the left menu 

<figure><img src="/files/rI51tkGGhyZ3FJSxgN9W" alt="" width="119"><figcaption></figcaption></figure>

* Now click on the Create Vector Search Index

<figure><img src="/files/Hz0xqCMUypPVc2yCrLy4" alt=""><figcaption></figcaption></figure>

* Select environment, db collection and vector search index name.

<figure><img src="/files/PtjlFyg2CSh9KGC39g1y" alt=""><figcaption></figcaption></figure>

* In the next step you need to define the mappings

<figure><img src="/files/9Sp9zXNsnEhuimRGhW73" alt=""><figcaption></figcaption></figure>

{% hint style="danger" %}
vector is a default field which is only there for the array of floats&#x20;
{% endhint %}

| Field                | Value                             | Example                                  |
| -------------------- | --------------------------------- | ---------------------------------------- |
| Type                 | vector \| field                   | vector                                   |
| Field Name           | fields in model                   | fields with type list of floats in model |
| Number Of Dimensions | number(1 - 4096)                  | 1023                                     |
| Similarity Function  | cosine \| euclidean \| dotProduct | cosine                                   |

{% hint style="danger" %}
Number Of Dimensions depends on the model which you have used to generate the vector embedings.
{% endhint %}

Once done with filling form then simply click on the create button on the bottom right of the form.