SST – Smart Sustainable Technologies

Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors
NVIDIA A100 80GB PCIe Gen4 Tensor Core GPU

Call for Price

NVIDIA A100 80GB PCIe is a high-performance data center GPU built on Ampere architecture, featuring 80GB HBM2e memory, third-generation Tensor Cores, and NVLink, delivering exceptional acceleration for AI, HPC, and data analytics workloads.

Category :

Category:

SKU :

SKU: P1001 SKU 210

Brand:

Additional information

Avalibility:

In Stock

Payment Methods:

Description

The NVIDIA A100 80GB PCIe Gen4 GPU is designed to accelerate AI training, inference, high-performance computing, and data analytics in enterprise data centers. Built on the Ampere GA100 architecture and equipped with 80GB of high-bandwidth HBM2e memory, it delivers up to 1.55 TB/s memory bandwidth and 312 TFLOPS of Tensor performance. Featuring Multi-Instance GPU (MIG) support for secure multi-tenant deployments, third-generation Tensor Cores for FP64 to INT8 workloads, and PCIe 4.0 / NVLink connectivity for scalable performance, the A100 80GB enables cutting-edge applications in machine learning, big data, and HPC environments.

Specifications

GPU Architecture NVIDIA Ampere (GA100)
CUDA Cores 6912
Tensor Cores 432 (3rd‑gen)
Memory Capacity 80 GB HBM2e
Memory Bandwidth ≈1.55 TB/s (5120‑bit bus)
FP64 9.7 TFLOPS
FP32 / TF32 19.5 TFLOPS / 19.5 TFLOPS
FP16 / BFLOAT16 312 TFLOPS
INT8 624 TOPS
PCIe Interface PCIe 4.0 ×16 (~64 GB/s)
NVLink Bridges Yes, 3× NVLink (600 GB/s bidirectional)
MIG Instances Up to 7 (~11 GB each)
TDP 300 W (configurable 150–300 W)
Cooling Passive heatsink (requires airflow)
Form Factor Full-height, dual-slot, 10.5″ FHFL
Power Connector 1× 8-pin auxiliary (CPU style)