• NVIDIA ampere architecture
• Third-generation tensor cores
• Multi-instance gpu (mig)
• Next-generation Nvlink
A100 is part of the complete NVIDIA data center solution that incorporates building blocks across hardware, networking, software, libraries, and optimized AI models and applications from NGC™.
Representing the most powerful end-to-end AI and HPC platform for data centers, it allows researchers to deliver real-world results and deploy solutions into production at scale.
Peak FP64 Tensor Core
Tensor Float 32 (TF32)
BFLOAT16 Tensor Core
FP16 Tensor Core
INT8 Tensor Core
Max Thermal Design Power (TDP)
Up to 7 MIGs @ 10GB
PCIe dual-slot air cooled or single-slot liquid cooled
NVIDIA® NVLink® Bridge for 2 GPUs: 600GB/s ** PCIe Gen4: 64GB/s
Server Options Partner and NVIDIACertified Systems™ with 1-8 GPUs