AMD Instinct MI355X AI Accelerator

The AMD Instinct MI355X is a flagship accelerator for data-centre AI training and HPC, built for dense clusters that need sustained throughput and faster turnaround on large models. Its CDNA 4 architecture and liquid-cooled deployment support more work per rack while helping maintain performance in tightly managed infrastructure.

AMD Instinct MI355X AI Accelerator
Submit an enquiry for pricing
SKU MI355X Categories , , Brand:
View all models in this series: AMD AI GPUs

Our Pricing Options | Shipping & Delivery Information | Customer FAQ

AMD Instinct MI355X Overview

As AI and HPC workloads grow denser, infrastructure teams need more throughput per rack without pushing power, cooling and space beyond practical limits.

High-Density Acceleration for Liquid-Cooled Platforms

The AMD Instinct MI355X is a GPU accelerator for large-scale AI training, inference and HPC in environments built for direct-liquid cooling. It is suited to dense clusters where accelerator throughput and rack efficiency matter more than conventional air-cooled deployment models.

Designed for Demanding Model and Simulation Workloads

With 5,000 TFLOPS of BF16 / FP16 performance, 288 GB of GPU memory and 8 TB/s of memory bandwidth, the platform supports large workloads that need sustained data movement and high compute density. Up to 8 GPUs per server helps teams concentrate performance into fewer systems.

Operational Headroom for Growth and Utilisation

This gives platform owners a cleaner path to higher utilisation, shorter training and inference queues, and better use of valuable data-centre space. It also supports large private AI and simulation workloads within tightly managed infrastructure.

Our team can help you assess the AMD Instinct MI355X for your liquid-cooled environment and map it to the workloads you need to run.

AMD Instinct MI355X Key Features​

AMD Instinct MI355X is a data center accelerator platform for AI and high-performance computing environments, designed to deliver dense compute capability and large-scale memory bandwidth for demanding training and inference workloads.

Scale AI Compute Density

Accelerator-Class Parallelism
The platform is designed for high-throughput GPU acceleration in large-scale AI and HPC deployments.

BF16 and FP16 Performance
It delivers up to 5000 TFLOPS of BF16 and FP16 performance to support compute-intensive model execution.

High-Capacity Memory Architecture
288 GB of GPU memory provides a large working set for data-heavy workloads and model processing.

High Bandwidth Data Movement
8 TB/s of memory bandwidth helps sustain rapid data access across accelerated workloads.

Support Demanding Workloads

Training and Inference Alignment
The platform is suited to AI workloads that need both dense compute throughput and broad memory capacity.

Large-Model Working Set Support
Its memory capacity helps accommodate large model states and intermediate data during execution.

Throughput-Oriented Execution
The architecture is built to keep accelerator resources supplied for sustained high-performance processing.

Speak to an AMD Server Specialist

When AI initiatives require high compute density and substantial memory bandwidth, our team can help map AMD Instinct MI355X into the right data center architecture.

AMD Instinct MI355X Technical Specifications

Full specifications for this model are listed below.

Additional information

Evaluating whether this is the right fit for your environment?

Our specialists are here to help assess compatibility, compare suitable alternatives, or talk through your configuration needs before committing to a solution.

Contact us today for a no-obligation chat.

AMD Instinct MI355X Deployment Scenarios

The AMD Instinct MI355X is built for dense accelerator deployments where AI and HPC workloads need maximum throughput in liquid-cooled infrastructure. It fits organisations that want to increase work completed per rack, reduce queue times, and keep very large models moving efficiently.

Liquid-Cooled Data Centre AI Racks

In high-density data centres, the pressure is on power, thermal limits and rack efficiency. MI355X supports operators that need more training and inference from each rack without expanding floor space, making it a strong choice for specialised liquid-cooled deployments.

Large Model Training and Iteration Clusters

Software teams working on large language and multimodal models need high throughput, large memory capacity and fast turnarounds between experiments. MI355X gives development clusters the headroom to train, fine-tune and test more aggressively within the same delivery window.

Private AI for Fraud, Risk and Quantitative Workloads

Finance teams running sensitive AI services need central control, strong performance and predictable execution on private datasets. MI355X supports tightly managed cluster deployments where fraud detection, risk modelling and quantitative analysis must stay fast and operationally contained.

Render Farms for Generative Media Production

Media and entertainment pipelines often face long render queues and tight delivery dates. MI355X suits dense production farms that need to clear heavy visual and video workloads faster while also supporting more complex AI-assisted content generation.

Simulation and Forecasting in Energy Operations

Energy and utilities teams use compute-heavy modelling to assess scenarios, optimise assets and improve forecasts. MI355X is a practical fit for liquid-cooled HPC and AI clusters where faster runs help teams work through more analysis before operational deadlines.

Planning an AMD Instinct MI355X Deployment?

Our team can help design and deploy AMD Instinct MI355X systems for AI training, inference, private finance workloads, media production and energy modelling, with practical support for integration, cooling, and deployment planning.

Tailored recommendations for your infrastructure

Below you’ll find alternative models, suitable software and services that pair with this solution – helping you to avoid compatibility issues, reduce support overhead and deploy with confidence.

Not sure where to start?

Not all deployments fit standard configurations. If you’re weighing up options or want a second opinion on your setup, our team is here to help with honest, straightforward advice backed by decades of vendor knowledge.

Speak to a specialist

Why choose us?

Expert advice

Get the right solution for your environment.

Access to leading vendors

Official partner across major technology brands.

Competitive pricing

Get the best value for your infrastructure.

Ongoing support

We’re here throughout the lifecycle.

Need help with this product?

Not sure which configuration, licence or support option is right for your requirements? Tell us what you need and our team can help with selection, compatibility, pricing and availability.

  • Advice from our technical specialists
  • No obligation
  • A quick response

Looking for a different model?

You may also like…

Need to define the right IT solution?

Alternatively, If you’re unsure whether this product fully meets your project’s needs, we’re here to help.
Speak to a Specialist
A group discussing IT solutions
Can’t find what you’re looking for? We have more products available - just reach out to us directly! Found a better price elsewhere? Let us know - we’re happy to discuss price matching.