FERRAMENTAS LINUX: AMD ROCm 6.4.1 Officially Released: Full RDNA4 GPU Support & Key Enhancements

quinta-feira, 22 de maio de 2025

AMD ROCm 6.4.1 Officially Released: Full RDNA4 GPU Support & Key Enhancements

 

Radeon

 AMD ROCm 6.4.1 is now available with official support for RDNA4 GPUs, including Radeon RX 9000 series. Discover new ROCm-DS for data science, installer updates, and compatibility with Ubuntu & RHEL.

ROCm 6.4.1: A Major Step Forward for AMD GPU Computing

Less than 24 hours after AMD’s Computex 2025 keynote hinted at improved ROCm support for consumer GPUs, AMD has officially launched ROCm 6.4.1—bringing critical upgrades for developers, data scientists, and high-performance computing users.

This release introduces formal support for RDNA4-based GPUs, including the Radeon RX 9000 series, alongside new features like ROCm-DS (Data Science toolkit), enhanced installers, and extensive documentation refinements.

ROCm

Key Features of ROCm 6.4.1

✅ Full RDNA4 GPU Support – Covers Radeon RX 9070 series, RX 9060 XT, and Radeon AI PRO R9700

✅ New ROCm-DS Component – Optimized for machine learning & AI workloads

✅ Expanded OS Compatibility – Works with Ubuntu 24.04.2, 22.04.5, and RHEL 9.4/9.5

✅ Improved Stability & Fixes – Numerous optimizations for better performance

Why This Release Matters for GPU Computing

With ROCm 6.4.1, AMD strengthens its position in AI, machine learning, and high-performance computing, directly competing with NVIDIA’s CUDA ecosystem. The inclusion of RDNA4 GPUs means broader accessibility for developers and researchers using consumer-grade hardware.

Supported GPUs & System Requirements

GPU SeriesSupported ModelsMinimum OS Requirement
RDNA4RX 9070, RX 9060 XT, Radeon AI PRO R9700Ubuntu 22.04.5 / RHEL 9.4
RDNA3RX 7900 SeriesUbuntu 20.04+ / RHEL 8.6+

Note: ROCm 6.4.1 does not yet include official support for Ryzen AI MAX "Strix Halo" hardware.

What’s Next for AMD ROCm?

AMD continues to expand ROCm’s capabilities, but day-one GPU support remains a challenge. Future updates may bring:

  • Broader consumer GPU compatibility

  • Enhanced AI/ML optimizations

  • Tighter integration with PyTorch & TensorFlow

For full details, visit AMD’s official ROCm documentation.


Nenhum comentário:

Postar um comentário