ROCm 6.1.3 released
AMD is officially launching its latest ROCm (Radeon Open Compute) platform, now featuring support for multiple graphics cards in clusters.
Earlier this month, AMD announced that the upcoming 6.1.3 update for ROCm will introduce several features requested by the community and developers. These features are important to those using AMD hardware for data science, complex compute workloads, machine learning inference, and rendering.
The ROCm update enhances support for more of AMD’s powerful RDNA3-based desktop graphics cards, including the Radeon RX gaming variants. The previous update added support for the RX 7900 GRE, while the latest update focuses on the newly launched PRO W7900 Dual-Slot. This special edition features the same specs as the existing 2.5-slot model but with a slimmer cooler design, allowing for more GPUs in multi-GPU configurations. These configurations are now supported by the compute platform.
Key feature enhancements in this release focus on improving compatibility, accessibility, and scalability, and include:
- Multi-GPU support to enable building scalable AI desktops for multi-serving, multi-user solutions.
- Beta-level support for Windows Subsystem for Linux, allowing these solutions to work with ROCm on a Windows OS-based system.
- TensorFlow Framework support offering more choice for AI development.
ROCm 6.1. 3, Source: AMD
Another important update is the support for WSL 2 (Windows Subsystem for Linux), enabling initial support for various Linux-based AI tools on Windows systems. However, it’s important to note that this support is still at the beta level and not yet native to Windows.
AMD has also announced that the TensorFlow framework has been qualified for ROCm. Alongside PyTorch and ONNX, this means that the ROCm library has been greatly improved for the most popular AI acceleration platforms.
AMD is committed to building a highly scalable and open ecosystem with ROCm software. With the latest release of ROCm 6.1 software for Radeon desktop GPUs, AMD empowers system builders to take full advantage of our enhanced solution stack to create on-prem systems that add powerful AI performance to any IT infrastructure, making them ideal for mission-critical or low-latency projects – and allowing organizations to keep their sensitive data in-house.
ROCm 6.1. 3, Source: AMD
It’s also worth noting that multi-GPU support in the latest ROCm update extends to up to four Radeon PRO or Radeon RX GPUs (x2 RX 7900XTX & W7900, x2 and x4 W7900 Dual-Slot). With a single Navi-31 based PRO GPU, users can access up to 48GB of memory per GPU. This capability enables the use of large language models with increased parameter requirements, such as LLaMA-3 with 70 billion parameters, according to AMD.
Source: AMD