Supported testing platforms
These images are a great way to quickly get started with Ubuntu Core and to explore Ubuntu Core’s features and evaluate its potential. Canonical produces official images for the following platforms:
| Platform / model name | Hardware / CPU | Cores / memory / storage | Image links |
|---|---|---|---|
Generic x86ubuntu-core-24 ubuntu-core-22 ubuntu-core-20 |
Full x86 32/64 bit | as defined | core-24-amd64 core-22-amd64 core-20-amd64 |
Raspberry Pi 5 ubuntu-core-24-pi ubuntu-core-22-piubuntu-core-20-pi |
Broadcom BCM2712 Cortex-A76 (ARM v8) 64-bit |
4 cores | 4/8 GB RAM no built-in storage |
core-24-pi-arm64 core-22-pi-arm64 core-22-pi-armhf core-20-pi-arm64 core-20-pi-armhf |
Raspberry Pi 4 ubuntu-core-24-pi ubuntu-core-22-piubuntu-core-20-pi |
Broadcom BCM2711 Cortex-A72 (ARM v8) 64-bit |
4 cores | 1/2/4/8 GB RAM no built-in storage |
core-24-pi-arm64 core-22-pi-armhf core-22-pi-arm64 core-20-pi-armhf core-20-pi-arm64 |
Raspberry Pi CM 3 ubuntu-core-22-piubuntu-core-20-pi |
Broadcom BCM2837 ARM Cortex-A53 64-bit |
4 cores | 1 GB RAM no built-in storage |
core-22-pi-armhf core-22-pi-arm64 core-20-pi-armhf core-20-pi-arm64 |
Raspberry Pi Zero 2 W and Raspberry Pi 3 ubuntu-core-22-piubuntu-core-20-pi |
Broadcom BCM2837 ARM Cortex-A53 64-bit |
4 cores | 0.5/1 GB RAM no built-in storage |
core-22-pi-armhf core-22-pi-arm64 core-20-pi-armhf core-20-pi-arm64 |
Raspberry Pi 2 ubuntu-core-22-piubuntu-core-20-pi |
Broadcom BCM2836 ARM Cortex-A7 |
4 cores | 1 GB RAM no built-in storage |
core-22-pi-armhf core-20-pi-armhf |
Ubuntu Core 24 (UC24) images are available from:
http://cdimage.ubuntu.com/ubuntu-core/24/stable/current/
Ubuntu Core 22 (UC22) images are available from:
http://cdimage.ubuntu.com/ubuntu-core/22/stable/current/
Ubuntu Core 20 (UC20) images are available from:
http://cdimage.ubuntu.com/ubuntu-core/20/stable/current/
Ubuntu community members also work with, and produce, images for other platforms and CPUs. Refer to these individual projects for more information on what other unofficial images might work for your use cases.