GPUDatacenterReleased Jun 28, 2021

NVIDIA A100 PCIe 80GB

80 GB HBM2e · 5120-bit · 6,912 cores

A PCI Express add-in card version of the Ampere data center accelerator with double the memory of the original 40 GB model and faster HBM2e, aimed at large AI models, recommender systems and HPC workloads limited by memory capacity and bandwidth. It keeps Multi-Instance GPU partitioning, and two cards can be paired with an NVLink bridge. Announced in June 2021.

General

Manufacturer: NVIDIA
Model: A100 PCIe 80GB
Segment: Datacenter
GPU chip: GA100
Architecture: Ampere
Release date: Jun 28, 2021
Price: Search on Amazon

Memory

Memory size: 80 GB
Memory type: HBM2e
Bus width: 5120-bit
Bandwidth: 1,935 GB/s
Memory clock: 1,512 MHz

Clocks & silicon

GPU clock: 1,065 MHz
Boost clock: 1,410 MHz
Shader units: 6,912 CUDA Cores
Tensor / AI cores: 432
RT cores: —
Transistors: 54.2 billion
Technology (nm): 7 nm
TSMC N7
TDP / TBP: 300 W

Interface & APIs

Bus interface: PCIe 4.0 x16
SLI / multi-GPU: NVLink bridge (600 GB/s, 2 GPUs)
DirectX: N/A (compute-only)
OpenGL: N/A (compute-only)
Vulkan: N/A (compute-only)
Other APIs: CUDA (compute capability 8.0)

AI support

Precisions: FP8FP16BF16INT8INT4TF32FP64
AI accelerator: Third-generation Tensor Cores (432)
Peak AI TOPS: 1,248 TOPS
Notes: Third-generation Tensor Cores supporting FP64, TF32, BF16, FP16, INT8, INT4 and binary; FP8 arrived with Hopper. Peak throughput with structured sparsity: INT8 1,248 TOPS, FP16 and BFLOAT16 624 TFLOPS, TF32 312 TFLOPS; dense rates are half. FP64 is 9.7 TFLOPS, FP64 Tensor Core and FP32 19.5 TFLOPS each. Multi-Instance GPU partitions the card into up to seven instances.

All NVIDIA GPUs →

Datacenter gpus from NVIDIA announced around the same time as the NVIDIA A100 PCIe 80GB. Add any of them to your compare list to see the specs side by side.

gpu