Loading the catalog…
NVIDIA A100 Tensor Core GPU (NVIDIA)
Blurred values are subscriber-only. Sign in to start a free trial — 2 full pages a day — or subscribe for unlimited access.
| Memory Bandwidth Tb S | 1.94 |
|---|---|
| Interconnect | NVIDIA NVLink Bridge for 2 GPUs: 600GB/s **; PCIe Gen4: 64GB/s |
| Fp32 Tflops | 19.5 |
| Multi Instance Support | Up to 7 MIGs @ 10GB |
| Memory Type | HBM2E |
| Gpu Memory Gb | 80 |
| Accelerator Form Factor | PCIe |
| Architecture | Ampere |
| Fp16 Bf16 Tflops Dense | 312 |
| Tdp W | 300 |
| Fp16 Bf16 Tflops Sparse | 624 |
| Int8 Tops Dense | 624 |
| Fp64 Tflops | |
| Int8 Tops Sparse | |
| Product Name | |
| Tf32 Tflops Dense | |
| Tf32 Tflops Sparse |
| Product family | NVIDIA A100 Tensor Core GPU |
|---|---|
| SKU | A100 80GB PCIE |
| Tagline | |
|---|---|
| Specifications form factors | |
| Tensor Float 32 (TF32) | |
| BFLOAT16 Tensor Core | |
| INT8 Tensor Core | |
| Server Options | |
| Footnote | |
| Tensor Cores generation | |
| Deep learning performance | |
| DRAM utilization efficiency | |
| Memory bandwidth (described) | |
| Max GPUs interconnected via NVSwitch | |
| Multi-Instance GPU (MIG) partitions | |
| Certification | |
| GPU-accelerated applications | |
| Product URL | |
| Document ID / date |
