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

# Overview

> A high level overview of the Metal Platform.

<img src="https://mintcdn.com/metal/JlNJrP-9OzWcypuS/images/diagram.png?fit=max&auto=format&n=JlNJrP-9OzWcypuS&q=85&s=e84817e2c05d57917f5ee53822986323" alt="Metal Platform" width="2166" height="1492" data-path="images/diagram.png" />

## Nouns

Critical resources of the Metal Platform.

| Resource | Description                                          |
| -------- | ---------------------------------------------------- |
| Document | An embedding resource representation                 |
| Index    | An indexed vector database of your *Embeddings*      |
| Org      | An organization                                      |
| Tuning   | A mechanism to fine-tune your *App* programatically. |

<img src="https://mintlify.s3.us-west-1.amazonaws.com/metal/images/resources.png" alt="Metal Platform" />

## Concepts

| Concept   | Description                                                      |
| --------- | ---------------------------------------------------------------- |
| Indexing  | Pushing data (raw text, documents, images, etc) into our system. |
| Searching | An operation to run semantic searches on your index.             |
| Tuning    | A mechanism to fine-tune your *App* programatically.             |

### Indexing

You can index various types of data into our system. We support the following data types:

* Image URLs (.jpg, .png, .gif, .bmp, .tiff)
* Text (string)
* Embeddings (number\[])

Checkout the [Index](/api-reference/documents/indexing) endpoint for more details.

### Searching

You can run semantic searches on your embedded data in a number of ways. We support the following search term types:

* Images (.jpg, .png, .gif, .bmp, .tiff)
* Text (string)

Checkout the [Search](/api-reference/core/search) endpoint for more details.

### Tuning

We offer a tuning mechanism to programatically improve the quality of your search results.
By labeling your data with the tune endpoint, you can improve the accuracy of your data by
generated custom embeddings for your specific use case.

Checkout the [Tune](/api-reference/core/tune) endpoint for more details.