At DigitalOcean, we’re committed to giving you even more options to power your AI/ML workloads. Today, we’re excited to announce that DigitalOcean customers now have access to AMD Instinct™ MI300X as DigitalOcean GPU Droplets.
AMD Instinct™ MI300X GPUs deliver leadership performance for accelerated high-performance computing (HPC) applications and the newly exploding demands of generative AI. Use-cases for these GPUs include large model training, fine-tuning, inference, and HPC.
With the AMD ROCm™ software platform, you can develop powerful HPC and AI production-ready systems faster than ever before. AMD Instinct™ MI300X’s large memory capacity allows it to hold models with hundreds of billions of parameters entirely in memory, reducing the need for model splitting across multiple GPUs.
By combining powerful AMD AI compute engines and DigitalOcean’s cloud technologies, we aim to empower the massive community of developers like you to integrate AI into your applications and support your most demanding AI workloads at scale.
Memory performance: High memory bandwidth (3.35 TB/s) and capacity (192 GB of HBM3) to efficiently handle larger models and datasets.
Value: Offered at a competitive price point ($1.99 GPU/hr/on-demand)
Customization: AMD Instinct™ MI300X is available both as single and eight GPU configurations and in bare metal configurations
Virtual instances to manage cost
Seamless integration with the broader DigitalOcean ecosystem, including access to our Kubernetes service
Pre-installed Python and Deep Learning software packages
HIPAA-eligibility and SOC 2 compliance (all GPU Droplets)
Later this year, we’ll also offer AMD Instinct™ MI325X GPUs, further expanding access to powerful and affordable GPU models. AMD Instinct™ MI325X GPU accelerators set new AI performance standards, delivering incredible performance and efficiency for training and inference.
Ready to get started? Spin up a GPU Droplet today.
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。