Available In Stock
NVIDIA H200 SXM
Tensor Core GPU
Built for memory-intensive AI workloads with expanded HBM3e capacity and improved throughput for trillion-token pipelines.
Architecture
Hopper
Process
4nm Custom
Transistors
80 Billion
Memory
141GB HBM3e
Technical Infrastructure
Comprehensive performance metrics and architectural specifications.
| Parameter | Value |
|---|---|
| Brand | NVIDIA |
| Products Status | Available |
| Application | Server |
| Architecture | Hopper |
| Memory | 141GB HBM3e |
| Memory Bandwidth | 4.8 TB/s |
| Tensor Performance | 1.9 PFLOPS FP8 |
| Interconnect | NVLink 4.0 |
| TDP | 700W |
| Form Factor | SXM |
| Use Case | Memory-intensive AI workloads |
| Deployment | Enterprise cluster |
*Lead time depends on regional stock allocation.
View Detailed Architecture Whitepaper open_in_newPrecision Engineering
Explore every detail of the hardware that powers the modern AI era.
HBM3e Density
Expanded memory footprint for large context and retrieval workloads.
Cluster Memory Scaling
Designed for memory-heavy training clusters and analytics nodes.
Hopper Platform
Continuity for enterprise deployments requiring mature software stacks.
Ready to Transform Your Infrastructure?
Our engineering team is ready to assist you in configuring the ideal deployment for your specific workload.
