HPE Cray XD675
Accelerate AI workloads with a game-changing, purpose-built architecture.
Run the largest AI models with fewer GPUs
Delivering up to 1307 teraflops of FP16 peak performance with 192GB of HBM3 per GPU, the AMD Instinct™ MI300X can run up to 80 billion parameter LLMs entirely in-memory.
Advance your AI workloads
Go beyond what’s possible and accelerate outcomes as you build and train your own models.
HPE and AMD are your trusted partners for the AI era
Profit from our supercomputing leadership with advanced technology and open architecture to secure your competitive advantage.
Unleash unimaginable performance and efficiency
Tap into the raw power of an 8-way HPC system from the supercomputing industry leader.
