# Platform Units

> For more general information on Platform Units consumption for our AI products, check the **Metering and charging logic** and **License tracking** sections.

## Overview

For more general information on Platform Units consumption for our AI products, check the **Metering and charging logic** and **License tracking** sections.

For specific details on Platform Units consumption for Process Mining, check out the [License](https://docs.uipath.com/process-mining/automation-cloud/latest/user-guide/pm-enabling-the-service-in-automation-cloud#license) page in the Process Mining guide.

You can also allocate and track Platform Units consumption at tenant level. Check the [Allocating licenses to tenants](https://docs.uipath.com/automation-cloud/automation-cloud/latest/admin-guide/allocating-licenses-to-tenants) page from the Automation Cloud<sup>TM</sup> guide for more details.

:::note
By default, each tenant is allocated 0 Platform Units, and all Platform Units are consumed from the account pool. If there are no Platform Units allocated to the tenant, Platform Units are consumed from the organization account pool. In case all Platform units are consumed from the tenant pool, then the admin needs to allocate more on that specific tenant.
:::

## Metering and charging logic

### General logic

This section contains specific information regarding Platform Units depending on the used activity, covering the cost for every AI product.

To calculate the overall consumption cost, the following formula is used:

`prediction cost` + `hardware cost` = `consumption cost`

For more information, check the following sections:

* Prediction cost
* Hardware cost

### Prediction cost

To calculate the prediction cost, the following formula is used:

`input size` x `unit cost of the model` = `prediction cost`

#### Input size

 <colgroup>
  <col/>
  <col/>
  <col/>
  <col/>
 </colgroup>
 
  
    Model 
    Input type 
    Input size 
    Computed input size 
  
 
 
  
    Document Understanding <sup> TM </sup> (UiPath and Customer-Managed third party) 
    Document 
    1 page 
    Number of pages in the input document 
  
  
    Communications Mining 
    JSON 
    1 message 
    Number of messages per mailbox or ticketing system 
  
  
    AI Computer Vision 
    Image 
    1 image 
    Always 1 
  
  
    Task Mining 
    Dataset 
    1 dataset 
    Always 1 
  
  
    GenAI Activities 
    String 
    String size limit is different for each model 
  
  
    Other models 
    JSON 
    2000 characters = 1 unit 
    Ceil(length(input)/2000) 
  
  
    File 
    5 MB = 1 unit 
    Ceil(size/5MB) 
  
  
    Files 
    5 MB = 1 unit 
    Ceil(sum(size(input))/5MB) 
  
 

#### Model used

| Model | When we charge | Platform Unit cost |
| --- | --- | --- |
| Document Understanding<sup>TM</sup> (UiPath and Customer-Managed third party) | Per prediction | For a list of all Document Understanding models, check the [Metering & Charging Logic](https://docs.uipath.com/document-understanding/automation-cloud/latest/classic-user-guide/metering-charging-logic) page from the Document Understanding guide. |
| AI Computer Vision | Per prediction | 0 |
| Models in preview (like UiPath Image Classification) | Per prediction | 0 |
| Task Mining | Per successful pipeline | 1000 |
| Communications Mining | Per message uploaded, modified, or predicted | 0.2 - for more information on **Communications Mining** charging logic, check the [official documentation](https://docs.uipath.com/communications-mining/automation-cloud/latest/developer-guide/introduction). |
| UiPath Light Text classifier | Per prediction | 0.04 |
| UiPath Multilingual classifier | Per prediction | 0.1 |
| UiPath Custom Named Entity Recognition | Per prediction | 0.1 |
| Open Source packages | Per prediction | 0.02 |
| GenAI Activities | Per execution | 0.2 - without Context grounding 0.4 - with Context grounding |

:::note
Except for Task Mining, running a pipeline or deploying an ML Skill only consumes Platform Units related to hardware consumption.
:::

### Hardware cost

The hardware cost at the time of deploying ML Skills is calculated as follows:

`replicas` x `resource cost`

The default replica count depends on the account type:

* Enterprise account: 2
* Other account types: 1

:::note
Availability is increased by increasing the number of replicas. High Availability (HA) is not guaranteed if user reduces the number of replicas to 1.
:::

| Hardware | Platform Units Cost |
| --- | --- |
| 0.5 CPU 2 GB RAM (default) | 0.2 Platform Unit / replica / hour |
| 1 CPU 4 GB RAM | 0.4 Platform Units / replica / hour |
| 2 CPU 8 GB RAM | 0.8 Platform Units / replica / hour |
| 4 CPU 16 GB RAM | 1.6 Platform Units / replica / hour |
| 6 CPU 24 GB RAM | 2.4 Platform Units / replica / hour |
| GPU | 4 Platform Units / replica / hour |

For hardware cost related to Pipelines, check the following table.

| Hardware | Platform Units Cost |
| --- | --- |
| CPU | 1.2 Platform Units / hour |
| GPU | 4 Platform Units / hour |

:::note
Any hour started is charged.
:::