Unsloth Expands Support for AMD GPUs in AI Model Training

What's Happening? Unsloth has announced official support for AMD hardware, allowing users to train, fine-tune, and deploy AI models on AMD GPUs. The open-source platform is compatible with Windows, WSL, and Linux, and supports a range of AMD GPUs, including Radeon RX 9000 and Instinct MI300 series.

AI & New Tech

SEE ALL

ARTnews

At the Art + Tech Summit, Christie’s Pins Its Future on AI Billionaires Who Aren’t Buying Art… Yet

Trendline

Artificial Intelligence Enhances Forensic Meteorology Amid Rising Weather Disasters

Discover daily

How One Software Update Triggered a Worldwide Technology Crisis

What is the story about?

What's Happening?

Unsloth has announced official support for AMD hardware, allowing users to train, fine-tune, and deploy AI models on AMD GPUs. The open-source platform is compatible with Windows, WSL, and Linux, and supports a range of AMD GPUs, including Radeon RX 9000

and Instinct MI300 series. Unsloth's collaboration with the community has enabled up to 2x faster training with 70% less VRAM usage without accuracy loss. The platform also offers remote access via Cloudflare HTTPS tunnels, enabling users to train models from mobile devices.

Why It's Important?

The expansion of Unsloth's support for AMD GPUs represents a significant development in the AI and machine learning landscape, providing more options for researchers and developers who rely on AMD hardware. This move could enhance the accessibility and efficiency of AI model training, particularly for those without access to high-end NVIDIA GPUs. By reducing VRAM usage and offering remote access capabilities, Unsloth makes AI development more feasible for a broader audience, potentially accelerating innovation and adoption in various industries.