NVIDIA Ampere GPUs

Realizes: High-throughput floating-point, tensor, and ray-tracing compute for AI/HPC workloads

NVIDIA's Ampere family (GA100, GA102) integrates second-generation ray tracing cores with third-generation tensor cores, powering the A100 data-center accelerator and GeForce RTX 3090 consumer card on a 7 nm FinFET substrate.

Examples

NVIDIA A100 (GA100) data-center GPU

The GA100-based A100 pairs second-gen ray tracing cores with third-gen tensor cores to accelerate matrix multiply, tensor fusion, and path-tracing workloads for AI training and HPC inference in datacenters.

MATRIX-MULTIPLY TENSOR-FMA RAY-TRACING AI-INFERENCE nanoseconds per cycle datacenter/hyperscale low energy per operation