# Requirements

## Hardware requirements

* CPU recommended: 3-4 threads as the base and 1-2 cores per worker on each machine.
* GPU Nvidia (consumer and professional) [computer capability](https://developer.nvidia.com/cuda-gpus) >= sm\_37 for optimal performance
* RAM GPU 4-6 GB for 1 worker depending of the resolution
* RAM > 16 GB
* 60 Go + local storage

## Software dependencies

* Linux Ubuntu >= 20.04 / CentOS >= 8 / Amazon Linux 2 Deep Learning AMI GPU 
* Windows 10 Pro, 11 Pro and server 2022 with Windows Subsystem for Linux (WSL)
* Install the following dependencies either using the provided script ([Linux](/2.4.7/blurit-on-premise/installation/linux-dependencies.md) only) or manually ([Linux](/2.4.7/blurit-on-premise/installation/linux-dependencies.md) or [Windows](/2.4.7/blurit-on-premise/installation/windows-dependencies-wsl.md)):
  * [nvidia drivers](https://www.nvidia.com/download/index.aspx) recommended >= 470
  * [docker ce](https://docs.docker.com/engine/install/) and [post install linux](https://.docker.com/engine/install/linux-postinstall/)
  * [docker desktop](https://docs.docker.com/desktop/install/windows-install/) for Windows WSL
  * docker-compose <= 1.29
  * [nvidia-docker2](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/latest/install-guide.html#setting-up-nvidia-container-toolkit) (set as default docker runtime)

## Blurit archive

* blurit\_stack.yml
* .env
* Blurit-op.postman\_collection.json
* README.md
* licence.txt
* login-registry.txt
* install\_dependencies.sh

###


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://doc-op.blurit.io/2.4.7/blurit-on-premise/requirements.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.