NVIDIA H100 SXM vs NVIDIA L4
How the NVIDIA H100 SXM and NVIDIA L4 compare across memory, compute, power and form factor, with the stronger figure highlighted on each row.
| Specification |
HOPPER
NVIDIA H100 SXM
View Product →
|
ADA LOVELACE
NVIDIA L4
View Product →
|
|---|---|---|
| Pricing | ||
| Price | – | $3,958.80 |
| Overview | ||
| Architecture | Hopper | Ada Lovelace |
| Memory | ||
| Memory | 80 GB HBM3 | 24 GB GDDR6 |
| Memory Bandwidth | 3.35 TB/s | 300 GB/s |
| Compute | ||
| FP32 | 67 TFLOPS | 30.3 TFLOPS |
| FP64 Tensor | 67 TFLOPS | – |
| TF32 Tensor | 989 TFLOPS | 120 TFLOPS (Sparse) |
| FP16/BF16 Tensor | 1,979 TFLOPS | 242 TFLOPS (Sparse) |
| FP8 Tensor | 3,958 TFLOPS | 485 TFLOPS (Sparse) |
| INT8 Tensor | 3,958 TOPS | 485 TOPS (Sparse) |
| Platform | ||
| Interface | Gen5 128 GB/s | PCIe Gen4 x16 |
| NVLink | 900 GB/s (4th Gen) | – |
| TDP | Up to 700W (configurable) | 72W |
| Form Factor | SXM | Single-slot, low-profile, passive cooling |
| Multi-Instance GPU (MIG) | Up to 7 instances (10GB each) | – |
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.