Skip to content
Launch imminent

Official date to be announced shortlyGPU offers available at launch; figures shown are indicative.

Priority access: sign up

NVIDIA Ampere architecture: A100, A30, RTX 30

Ampere is the NVIDIA architecture of the A100, A30, A10 and RTX 30 series, which introduced TF32, BF16 Tensor Cores and MIG.

Glossary3 sections1 min read

What is Ampere?

Ampere is the NVIDIA architecture behind the A100, A30 and A10 data center GPUs, RTX A-series workstation cards and the GeForce RTX 30 series. It introduced TF32 and BF16 Tensor Core support and MIG.

Why it matters when renting a GPU

Ampere GPUs, especially the A100, remain widely available and cost-effective for training and inference without FP8.

They lack FP8 Tensor Cores, so FP8-optimized software gains nothing on them.

Worked example

EverXCloud describes 28 Ampere models. The A100 SXM 80GB has 80 GB of HBM2e at 2,039 GB/s and 312 dense BF16 TFLOPS.

Frequently asked questions

Is the A100 still worth renting?
Often yes for BF16 training and inference, where its price per GPU-hour can be attractive. Compare against H100 for FP8 workloads.

Be ready on launch day

Create your account now: you can top up your balance and launch a GPU as soon as the market opens.