NVIDIA A100 GPU accelerator (PCIe card)

Graphics processor Released May 14, 2020

Memory bus: 5,120 bit^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Ampere

GPU configurations

Reference configuration
Rated power
250 W
Power specification
TDP

Memory & storage

Memory bus: 5,120 bit^*

Power specifications

Thermal design power: 250 W

System configuration

Detailed specifications

Specifications 1 · 40 fields

GPU architecture — NVIDIA A100 40GB PCIe — 40 GB HBM2: Ampere^*
GPU architecture — NVIDIA A100 80GB PCIe — 80 GB HBM2e: Ampere^*
Theoretical peak: BF16 tensor dense per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 312^* TFLOPS
Theoretical peak: BF16 tensor dense per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 312^* TFLOPS
Theoretical peak: FP16 tensor dense per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 312^* TFLOPS
Theoretical peak: FP16 tensor dense per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 312^* TFLOPS
Theoretical peak: FP32 non-tensor dense per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 19.5^* TFLOPS
Theoretical peak: FP32 non-tensor dense per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 19.5^* TFLOPS
Native hardware support: FP4 tensor — NVIDIA A100 40GB PCIe — 40 GB HBM2: No^*
Native hardware support: FP4 tensor — NVIDIA A100 80GB PCIe — 80 GB HBM2e: No^*
Theoretical peak: FP64 non-tensor dense per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 9.7^* TFLOPS
Theoretical peak: FP64 non-tensor dense per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 9.7^* TFLOPS
Theoretical peak: FP64 tensor dense per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 19.5^* TFLOPS
Theoretical peak: FP64 tensor dense per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 19.5^* TFLOPS
Native hardware support: FP6 tensor — NVIDIA A100 40GB PCIe — 40 GB HBM2: No^*
Native hardware support: FP6 tensor — NVIDIA A100 80GB PCIe — 80 GB HBM2e: No^*
Native hardware support: FP8 tensor — NVIDIA A100 80GB PCIe — 80 GB HBM2e: No^*
Native hardware support: FP8 tensor — NVIDIA A100 40GB PCIe — 40 GB HBM2: No^*
Theoretical peak: INT8 tensor dense per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 624^* TOPS
Theoretical peak: INT8 tensor dense per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 624^* TOPS
Theoretical peak: TF32 tensor dense per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 156^* TFLOPS
Theoretical peak: TF32 tensor dense per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 156^* TFLOPS
Physical configuration — NVIDIA A100 80GB PCIe — 80 GB HBM2e: PCIe, dual-slot air-cooled or single-slot liquid-cooled^*
Physical configuration — NVIDIA A100 40GB PCIe — 40 GB HBM2: PCIe, dual-slot FHFL^*
GPU interconnect bandwidth (NVLink 3) — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 600^* GB/s
GPU interconnect bandwidth (NVLink 3) — NVIDIA A100 40GB PCIe — 40 GB HBM2: 600^* GB/s
PCIe interface declaration — NVIDIA A100 40GB PCIe — 40 GB HBM2: PCIe Gen4 x16^*
PCIe interface declaration — NVIDIA A100 80GB PCIe — 80 GB HBM2e: PCIe Gen4^*
Maximum board power — NVIDIA A100 40GB PCIe — 40 GB HBM2: 250^* W
Maximum board power — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 300^* W
Peak memory bandwidth per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 1,935^* GB/s
Peak memory bandwidth per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 1,555^* GB/s
Memory interface width per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 5,120^* bit
Memory interface width per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 5,120^* bit
Memory capacity per GPU — NVIDIA A100 40GB PCIe — 40 GB HBM2: 40^* GB
Memory capacity per GPU — NVIDIA A100 80GB PCIe — 80 GB HBM2e: 80^* GB
Memory technology — NVIDIA A100 80GB PCIe — 80 GB HBM2e: HBM2e^*
Memory technology — NVIDIA A100 40GB PCIe — 40 GB HBM2: HBM2^*
GPU model — NVIDIA A100 40GB PCIe — 40 GB HBM2: NVIDIA A100 40GB PCIe^*
GPU model — NVIDIA A100 80GB PCIe — 80 GB HBM2e: NVIDIA A100 80GB PCIe^*

Specifications 2 · 3 fields

model

A100 GPU accelerator (PCIe card)^*

Scope and conditions

source_native_row_context

model: A100 GPU accelerator (PCIe card)^*

tdp watts

250.0^* W

Scope and conditions

source_native_row_context

model: A100 GPU accelerator (PCIe card)^*

manufacturer

Nvidia^*

Scope and conditions

source_native_row_context

model: A100 GPU accelerator (PCIe card)^*

* Provisional data