AMD Instinct MI350X vs NVIDIA H200 SXM
How the AMD Instinct MI350X and NVIDIA H200 SXM compare across memory, compute, power and form factor, with the stronger figure highlighted on each row. Vendor-specific figures (CUDA vs stream processors, Tensor vs matrix cores) are listed on their own rows.
| Specification | CDNA 4 AMD Instinct MI350X NVIDIA alternative: NVIDIA B200 → |  HOPPER NVIDIA H200 SXM View Product → |
|---|---|---|
| Pricing | ||
| Price | Not sold by NVNexus | – |
| Overview | ||
| Architecture | AMD CDNA 4 | Hopper |
| Memory | ||
| Memory | 288 GB HBM3E | 141 GB HBM3e |
| Memory Bandwidth | 8 TB/s | 4.8 TB/s |
| Memory Interface | 8192-bit | – |
| Compute | ||
| Stream Processors (AMD) | 16,384 (256 CUs) | – |
| Matrix Cores / AI Accelerators (AMD) | 1,024 | – |
| FP32 | 144.2 TFLOPS | 67 TFLOPS |
| FP64 Tensor | 72.1 TFLOPS | – |
| FP16/BF16 Tensor | 2.3 PFLOPS / 4.6 (sparse) TFLOPS | 1,979 TFLOPS |
| FP8 Tensor | 4.6 PFLOPS / 9.2 (sparse) TFLOPS | 3,958 TFLOPS |
| FP4 Tensor | 9.2 PFLOPS TFLOPS | – |
| INT8 Tensor | – | 3,958 TOPS |
| Platform | ||
| Interface | PCIe Gen5 x16 | Gen5 128 GB/s |
| NVLink / Infinity Fabric | Infinity Fabric, 1,075 GB/s aggregate | 900 GB/s (4th Gen) |
| TDP | 1000W | Up to 700W (configurable) |
| Form Factor | OAM module (8-GPU platform) | SXM |
| Multi-Instance GPU (MIG) | – | Up to 7 instances |
| Cooling | Air cooled | – |
| Transistors | 185 billion | – |
Not sure which fits?
Talk to an Expert → Talk it through with a specialist
Tell us your workload and we will recommend the right GPU, configuration, and sourcing options.
More Comparisons
- AMD Instinct MI350X vs AMD Instinct MI355X →
- AMD Instinct MI350X vs AMD Instinct MI210 →
- AMD Instinct MI350X vs AMD Instinct MI300X →
- AMD Instinct MI350X vs AMD Instinct MI325X →
- AMD Instinct MI350X vs NVIDIA B200 →
- AMD Instinct MI350X vs NVIDIA A10 →
- NVIDIA H200 SXM vs NVIDIA H100 NVL →
- NVIDIA H200 SXM vs NVIDIA H100 SXM →
- NVIDIA H200 SXM vs NVIDIA H200 NVL →
- NVIDIA H200 SXM vs NVIDIA A10 →