Professional NVIDIA GPUs for AI, ML and visualisation

A powerful server-class GPU for AI and professional visualisation — accelerating AI, scientific computing, graphics and video production.

• NVIDIA Blackwell architecture
• 96 GB GDDR7 with ECC
• Memory bandwidth up to 1.6 TB/s
• PCI Express 5.0 x16
• Max power: 600 W

Available to order

The most powerful universal GPU for data centres: breakthrough acceleration for LLM inference and training, graphics and video — a leading platform for multimodal generative AI.

• NVIDIA Ada Lovelace architecture
• 48 GB GDDR6 with ECC
• Memory bandwidth up to 864 GB/s
• Max power: 300 W

Available to order

Exceptional performance for generative AI and HPC: the first GPU with HBM3e brings greater capacity and faster memory for LLMs and scientific computing.

• NVIDIA Hopper architecture
• 141 GB HBM3e memory
• Max LLM inference performance
• Max power: up to 600 W

Available to order

Next-generation RT, Tensor and CUDA cores with 24 GB of GDDR7 — maximum performance in a compact form factor.

• NVIDIA Blackwell architecture
• 24 GB GDDR7 with ECC
• Max power: 70 W
• Compact single-slot design

Available to order

Breakthrough performance in a compact, energy-efficient design: 16 GB of ultra-fast GDDR7 for demanding AI workflows and professional graphics.

• NVIDIA Blackwell architecture
• 16 GB GDDR7 with ECC
• Max power: 70 W
• Energy-efficient design

Available to order

A new class of computer for AI development: prototype, fine-tune and run inference on large models locally — up to 1 petaFLOP in a compact form factor.

• NVIDIA Grace Blackwell architecture
• DGX OS operating system
• 128 GB LPDDR5x unified memory
• Read our real-world benchmarks

Available to order