# Quick start guide Welcome to AiQu! On this page, you'll find quick-start videos, an overview of AiQu's key features, and explanations of some basic concepts to help you get started quickly and easily. ## Quick start videos
To get you started right away, we have prepared a few short videos showing you the key features of AiQu.
🔹Quick deploy of a chatbot Watch Video
🔹Quick deploy of Jupyter Notebook Watch Video
🔹Persistent storage and accessing files Watch Video
We hope these videos help you get up and running quickly! Can't get enough of our video content? We’ve got plenty more video tutorials waiting for you on our [Video Guides page](https://docs.aiqu.ai/include/user_docs/video_guides/video_guides.html)! If you have any questions, feel free to explore our [FAQ section](https://docs.aiqu.ai/include/user_docs/faq.html). If you don't find the answers you're looking for, we're always here to help – just reach out to us!
## What is AiQu? AiQu is an orchestration platform designed to manage and optimize GPU-based workloads, particularly for AI and machine learning applications. It efficiently allocates GPU resources to different tasks. To streamline development, AiQu provides quick-deploy containers with prebuilt environments, including common frameworks and dependencies needed for AI workloads. Users can open ports to their jobs and connect to these environments, for example using SSH, the API, or even the AiQu terminal, making it simple to integrate AiQu into existing workflows. ### Key Features - **User friendly interface** - AiQu offers an intuitive and easy-to-use interface, ensuring a seamless experience for both beginners and advanced users. - **Automatic Workload Management** – AiQu handles scheduling and distributing GPU workloads, so resources are used efficiently without manual setup. - **Custom Resource Allocation** – You have full control over how many GPUs and CPUs to allocate for each task. Real-time and historical metrics help you monitor usage and optimize performance. - **Advanced Access Control & Queuing** – Organize work with projects, groups, and queues, managing access and permissions to ensure efficient collaboration and resource allocation. - **Prebuilt Environments** – Ready-to-use containers come with essential AI tools and frameworks, so you can start working immediately. - **Easy Access** – Connect to your environments using SSH or the API, making it simple to integrate AiQu into your workflow. - **Flexible Deployment** – Use AiQu as a cloud service with token-based access or install it on-premises to keep full control over your infrastructure. - **Secure and Isolated** – Users are separated from underlying compute resources, ensuring a safe and controlled environment for AI workloads. - **Edge Inference Support** – AiQu enables performing inference on edge devices like cameras for quality assessment, allowing real-time decision-making without relying on centralized servers. ## What is AiQu Trial? AiQu Trial provides time-limited access to AiQu with a simplified feature set, allowing users to explore its capabilities and experience how easy it is to use. It’s designed for quick hands-on testing, letting you deploy workloads, interact with the API, and see how AiQu manages resources. ## Basic concepts ### GPU vs CPU A **Graphics Processing Unit (GPU)** is a specialized processor designed for parallel processing, commonly used in graphics rendering and AI workloads. A **Central Processing Unit (CPU)** is the general-purpose processor that handles a computer’s main instructions and sequential tasks. So to explain the difference, CPUs are optimized for executing a few complex tasks efficiently, while GPUs excel at handling many simple tasks simultaneously. For example, a CPU is well-suited for running operating systems, managing databases, and handling application logic that requires sequential execution. On the other hand, a GPU is ideal for training deep learning models, rendering real-time graphics, processing high-resolution video, and performing large-scale scientific simulations. ### Job A job is a unit of work executed in AiQu. Starting a job means launching a Docker container that occupies the designated resources of the machine it runs on, whether that’s a physical machine or a virtual machine (VM). The container runs the specified scripts, and you can connect to it via terminal to interact with and manage the process. Learn more on our [Jobs page](https://docs.aiqu.ai/include/user_docs/jobs.html). ### Ports Ports enable network communication between your job and external systems, allowing secure access via SSH, APIs, or other protocols. Learn more on our [Exposing ports page](https://docs.aiqu.ai/include/user_docs/ports.html) ### Quick deploy Quick deploy allows you to instantly start a job using pre-configured environment images, which come with all necessary libraries and dependencies already installed. This means that when you log into your job, your environment is ready to go, saving you the time and effort of manual setup. See our tutorials on a [chatbot quick deploy](https://docs.aiqu.ai/include/user_docs/chatbot_quickdeploy.html) and [Matlab quick deploy](https://docs.aiqu.ai/include/user_docs/matlab.html)