Skip to content
NVNEXUS
info@nvnexus.com
Talk to an Expert →
GPU Comparison

AMD Instinct MI300X vs NVIDIA H100 SXM

How the AMD Instinct MI300X and NVIDIA H100 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 3 AMD Instinct MI300X NVIDIA alternative: NVIDIA H100 SXM → NVIDIA H100 SXM HOPPER NVIDIA H100 SXM View Product →
Pricing
Price Not sold by NVNexus –
Overview
Architecture AMD CDNA 3 Hopper
Memory
Memory 192 GB HBM3 80 GB HBM3
Memory Bandwidth 5.3 TB/s 3.35 TB/s
Memory Interface 8192-bit –
Compute
Stream Processors (AMD) 19,456 (304 CUs) –
Matrix Cores / AI Accelerators (AMD) 1,216 –
FP32 163.4 TFLOPS (matrix) 67 TFLOPS
FP64 Tensor 163.4 TFLOPS (matrix) 67 TFLOPS
TF32 Tensor 653.7 TFLOPS / 1,307 (sparse) 989 TFLOPS
FP16/BF16 Tensor 1,307 TFLOPS / 2,615 (sparse) 1,979 TFLOPS
FP8 Tensor 2,615 TFLOPS / 5,230 (sparse) 3,958 TFLOPS
INT8 Tensor 2,615 TOPS / 5,230 (sparse) 3,958 TOPS
Platform
Interface PCIe Gen5 x16 Gen5 128 GB/s
NVLink / Infinity Fabric Infinity Fabric, 896 GB/s aggregate 900 GB/s (4th Gen)
TDP 750W Up to 700W (configurable)
Form Factor OAM module (8-GPU platform) SXM
Multi-Instance GPU (MIG) – Up to 7 instances (10GB each)
Cooling Passive (chassis airflow) –
Transistors 153 billion –
Not sure which fits?

Talk it through with a specialist

Tell us your workload and we will recommend the right GPU, configuration, and sourcing options.

Talk to an Expert →

More Comparisons