NVIDIA A100 Enterprise 40GB/80GB PCIe
The NVIDIA A100 Tensor Core GPU redefines performance at every scale, driving the world’s most powerful elastic data centers for AI, data analytics, and high-performance computing (HPC) applications.
As the cornerstone of the NVIDIA data center platform, the A100 delivers up to 20x the performance of the previous NVIDIA Volta generation. With Multi-Instance GPU (MIG) technology, it can scale efficiently or be partitioned into seven isolated GPU instances, offering a unified platform that adapts dynamically to changing workload demands in elastic data center environments.
Contact us for pricing
Tech Specs
CUDA Cores | 6912 |
|---|---|
Streaming Multiprocessors | 108 |
Tensor Cores | Gen 3 | 432 |
GPU Memory | 40 GB or 80 GB HBM2e ECC on by Default |
Memory Interface | 5120-bit |
Memory Bandwidth | 1555 GB/s |
NVLink | 2-Way, 2-Slot, 600 GB/s Bidirectional |
MIG (Multi-Instance GPU) Support | Yes, up to 7 GPU Instances |
FP64 | 9.7 TFLOPS |
FP64 Tensor Core | 19.5 TFLOPS |
FP32 | 19.5 TFLOPS |
TF32 Tensor Core | 156 TFLOPS | 312 TFLOPS* |
BFLOAT16 Tensor Core | 312 TFLOPS | 624 TFLOPS* |
FP16 Tensor Core | 312 TFLOPS | 624 TFLOPS* |
INT8 Tensor Core | 624 TOPS | 1248 TOPS* |
Thermal Solutions | Passive |
vGPU Support | NVIDIA Virtual Compute Server (vCS) |
System Interface | PCIE 4.0 x16 |
Update 06.01.2024: Production of this product ended in February 2024 and it is now designated as EOL (end-of-life). Remaining stock is limited. NVIDIA recommends the L40/L40S series as a replacement, along with the more budget-friendly RTX A6000 ADA or the high-performance, premium-priced H100. A direct successor has not yet been announced.