As enterprises accelerate the deployment of AI applications, the need for intelligent, scalable hardware has never been greater. The HPE ProLiant DL384 Gen12 rises to the challenge, delivering groundbreaking performance for AI workloads.
Cutting-Edge Architecture for Next-Level AI Workloads
This new 2U rack server is pioneering as HPE’s first to incorporate the NVIDIA GH200 Grace Hopper NVL2 Superchip, bringing together two Grace CPU clusters and two Hopper GPUs in one unified architecture. It packs up to 8 petaflops of compute power, ideal for heavy-duty tasks such as fine-tuning large language models and inference using Retrieval-Augmented Generation (RAG).
Unmatched Memory Capacity & Speed
The DL384 Gen12 offers a staggering 1.2 TB of coherent memory across two GH200 Superchips, coupled with high-bandwidth connectivity. This design provides 3.5 times more GPU memory and delivers 3 times the memory bandwidth compared to systems with NVIDIA H100 GPUs.
Optimized for Enterprise AI & Hybrid Operations
Designed for agility, the server supports NVIDIA InfiniBand, Ethernet, and BlueField adapters, ensuring low latency and high throughput for AI data pipelines. It also builds on the HPE ecosystem through iLO management with Silicon Root of Trust, offering enterprise-grade security and streamlined infrastructure control.

Proven to Lead in AI Benchmarking
In MLPerf inference benchmarks, the DL384 Gen12 has secured multiple #1 performance accolades, showcasing its leadership in delivering efficient, high-speed AI computations.
Why It Matters
With its unified memory architecture and breakthrough performance, the HPE ProLiant DL384 Gen12 empowers organizations to train and deploy large-scale AI models with greater speed and efficiency—while maintaining simplicity, security, and manageability. Its advanced design makes it an ideal choice for enterprises looking to scale AI initiatives and modernize their infrastructure.
Curious to learn more? For a deeper dive into its specifications and how it unlocks new possibilities in AI infrastructure, head over to our detailed blog: HPE ProLiant DL384 Gen12 Server