ROCm-LS 25.09 release notes#
2 min read time
The release notes provide a summary of notable changes since the previous ROCm-LS release.
This early access release serves as a preview and is not intended for production workloads.
ROCm-LS release highlights#
The following are notable new features and improvements in ROCm-LS 25.09 since the release of 25.06.
Introducing MONAI for AMD ROCm#
MONAI (Medical Open Network for AI) is a PyTorch-based, open-source framework designed for deep learning in healthcare imaging. AMD now provides ROCm-enabled MONAI builds optimized for AMD GPUs as a ROCm-LS component, which enables scalable medical AI workflows on AMD hardware.
MONAI for AMD ROCm offers the following advantages:
GPU acceleration on AMD Instinct MI300X GPUs.
API compatibility with upstream MONAI without requiring any code changes.
Validation with hipCIM image processing workflows.
For more information, see MONAI for AMD ROCm documentation
ROCm-LS components#
The following table lists the versions of ROCm-LS components for ROCm-LS 25.09, including any version changes from 25.06 to 25.09. Click the GitHub icon to go to the component’s source code.
| Category | Component name | Version | Source code |
|---|---|---|---|
| Imaging | hipCIM | 1.0.00 | |
| AI/ML | MONAI for AMD ROCm | 1.0.0 |
Note
The hipCIM version remains unchanged in this release.