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

CDNA 3 Architecture
Hopper Architecture
| Specification | AMD MI300X | NVIDIA H200 |
|---|---|---|
| Architecture | CDNA 3 | Hopper |
| Release Year | 2023 | 2023 |
| VRAM | 192 GB+36.2% | 141 GB |
| Memory Type | HBM3 | HBM3e |
| Memory Bandwidth | 5300 GB/s+10.4% | 4800 GB/s |
| FP32 Performance | 163 TFLOPS+143.3% | 67 TFLOPS |
| FP16 Performance | 326 TFLOPS+143.3% | 134 TFLOPS |
| INT8 Performance | 2610 TOPS | 2680 TOPS+2.6% |
| Tensor Cores | N/A | 16896 |
| CUDA Cores | 19456 | 16896 |
| TDP | 750W | 700W |
| Form Factor | OAM | SXM |
| NVLink Support | No | Yes |
| Avg. Price/Hour | $2.5+11.1% | $2.25 |
Single-precision floating-point performance for general compute workloads
AMD MI300X is 143.3% faster
Half-precision performance optimized for deep learning training
AMD MI300X is 143.3% faster
Integer performance for efficient model inference and deployment
NVIDIA H200 is 2.6% faster
Data transfer speed between GPU and memory
AMD MI300X is 10.4% faster
AMD MI300X
AMD MI300X
NVIDIA H200
AMD MI300X

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