Comprehensive comparison for AI, machine learning, and high-performance computing workloads.

CDNA 2 Architecture
Hopper Architecture
| Specification | AMD MI210 | NVIDIA H200 |
|---|---|---|
| Architecture | CDNA 2 | Hopper |
| Release Year | 2021 | 2023 |
| VRAM | 64 GB | 141 GB+54.6% |
| Memory Type | HBM2e | HBM3e |
| Memory Bandwidth | 1638 GB/s | 4800 GB/s+65.9% |
| FP32 Performance | 23.9 TFLOPS | 67 TFLOPS+64.3% |
| FP16 Performance | 47.9 TFLOPS | 134 TFLOPS+64.3% |
| INT8 Performance | 192 TOPS | 2680 TOPS+92.8% |
| Tensor Cores | N/A | 16896 |
| CUDA Cores | 6656 | 16896 |
| TDP | 300W | 700W |
| Form Factor | PCIe | SXM |
| NVLink Support | No | Yes |
| Avg. Price/Hour | $0.82 | $2.25+63.6% |
Single-precision floating-point performance for general compute workloads
NVIDIA H200 is 64.3% faster
Half-precision performance optimized for deep learning training
NVIDIA H200 is 64.3% faster
Integer performance for efficient model inference and deployment
NVIDIA H200 is 92.8% faster
Data transfer speed between GPU and memory
NVIDIA H200 is 65.9% faster
NVIDIA H200
NVIDIA H200
AMD MI210
NVIDIA H200

Enterprise-grade infrastructure
Get a custom quote in 24 hours for reserved GPU clusters with high-speed interconnect, any region, any GPU model, and any number of GPUs you need.
Any GPU
Choose your hardware
Any Quantity
Scale as needed
Any Region
Global availability
Interconnect
High-speed networking
Go from comparison to running workload in under 60 seconds. No complex setup required.
Only pay for what you use. Stop instances anytime. No hidden fees or long-term commitments.
Enterprise-grade infrastructure with 99.9% uptime. Trusted by AI teams worldwide.
Explore more GPU comparisons to find the perfect match for your workload