System Requirements
System requirements for installing and running Nara software.
Operating System and Architecture Support
Please see below for the supported OS and architectures for Nara.
| OS | Distribution/ Version | Architecture | CPU | GPU (CUDA) | Note | Installation guide | Download Link |
|---|---|---|---|---|---|---|---|
| Windows | Windows 11 | x86_64 | ✓ | ✓ | Link | Download | |
| Linux | Ubuntu 22.04 | x86_64 | ✓ | ✓ | Link | Coming soon | |
| Linux | Ubuntu 24.04 | x86_64 | ✓ | ✓ | Link | Download | |
| Linux | Jetpack 6.2 | aarch64 | ✗ | ✓ | Based on Ubuntu 22.04 | Link | Download |
| Linux | Raspberry Pi OS 5 | aarch64 | ✓ | ✗ | Based on Debian 12 (Bookworm) | Link | Download |
| Linux | Raspberry Pi OS 6 | aarch64 | ✓ | ✗ | Based on Debian 13 (Trixie) | Link | Download |
Info
x86_64 is also known as AMD64 or x64.
aarch64 is also known as ARM64.
Minimum Requirements
| Hardware | Requirement |
|---|---|
| Processor | 1 GHz, 4 cores |
| Memory (RAM) | 2 GB |
| Storage Space | 8 GB |
| Storage Space (CUDA) | +5 GB for CUDA build |
Tip
Nara utilised CPU concurrency to run multiple tasks at the same time during the flow run. The higher number of CPU cores reduces the change of CPU bottleneck. 8 CPU cores is recommended.
Hardware Acceleration Support
Nara utilises hardware acceleration to speed up AI model inference and other processing tasks. Below are the supported hardware acceleration modes:
Nvidia GPU Support
Nara supports NVIDIA GPU with CUDA capability.
- Compute Capability: 7.5, 8.0, 8.6, 8.7, 8.9, 9.0, 10.0, 12.0, please refer to NVIDIA CUDA GPUs for more information.
Info
The higher amount of VRAM allows you to run a larger AI model or more number of models at the same time. At least 8GB of VRAM is recommended when using vision models.