At Computex 2025 in Taipei, NVIDIA CEO Jensen Huang unveiled the NVIDIA DGX H200, a state-of-the-art AI supercomputer designed to meet the growing demands of large-scale AI and high-performance computing (HPC) workloads. With a price tag exceeding $300,000, the DGX H200 is tailored for enterprises and research institutions seeking unparalleled computational power.

🔍 Key Features of the NVIDIA DGX H200

1. Unprecedented GPU Memory and Performance

• 8x NVIDIA H200 Tensor Core GPUs

• 1,128GB total GPU memory

• 32 petaFLOPS FP8 performance

The DGX H200 boasts eight NVIDIA H200 GPUs, each equipped with 141GB of HBM3e memory, delivering a combined 1,128GB. This configuration ensures exceptional performance for training and inference of large language models and complex simulations.

2. Advanced Interconnect and Networking

• 18x NVIDIA NVLink connections per GPU

• 4x NVIDIA NVSwitches

• 7.2TB/s bidirectional GPU-to-GPU bandwidth

• 10x NVIDIA ConnectX-7 400Gb/s network interfaces

The system’s high-speed interconnects and networking capabilities facilitate rapid data transfer and efficient scaling across multiple DGX systems.

3. Robust CPU and Memory Configuration

• Dual Intel Xeon Platinum 8480C processors (112 cores total)

• 2TB system memory

Complementing the powerful GPUs, the DGX H200 includes dual Intel CPUs and substantial system memory to handle data preprocessing and other CPU-intensive tasks.

4. High-Speed Storage

• 30TB NVMe SSD storage

The ample storage ensures quick access to large datasets, crucial for AI training and inference workloads.

💼 Enterprise Applications

The DGX H200 is engineered for:

• Training and inference of large language models (LLMs)

• High-performance computing tasks

• Scientific research and simulations

• Enterprise AI deployments

• Its architecture supports seamless integration into existing data centers and can be scaled to meet increasing computational demands.

💲 Pricing and Availability

The NVIDIA DGX H200 is priced between $300,000 and $500,000, depending on configuration and vendor. It’s available for purchase through NVIDIA’s partners and certified resellers.

📈 Conclusion

With the introduction of the DGX H200, NVIDIA continues to lead in providing cutting-edge solutions for AI and HPC workloads. Its unparalleled performance and scalability make it a valuable asset for organizations aiming to advance their AI capabilities.

Also Read:

Google Unveils “Flow”: A Game-Changing AI Tool for Cinematic Video Creation

Apple iPhone designer Sir Jony Ive joins ChatGPT-maker OpenAI