How much does the NVIDIA A100 80GB cost per hour to rent?

On Runcrate the NVIDIA A100 80GB starts at $1.73/hr per GPU on demand today, up to $4.26/hr depending on region and configuration. Billed by the second on most GPUs (a few bill by the minute), with no commitment. Reserved capacity is quoted on request.

Can I rent the NVIDIA A100 80GB on demand right now?

Yes. The NVIDIA A100 80GB is on demand in 6 regions today, billed by the second on most GPUs (a few bill by the minute) with no contract.

A100 vs H100: when is A100 still the right choice?

A100 costs less per hour than H100 and is still excellent for inference up to 13B and QLoRA fine-tuning up to 70B. H100 wins for FP8 workloads, transformer training at scale, and any modern PyTorch flash-attention path. For budget-constrained ML, A100 is the value pick.

A100 80GB or A100 40GB?

Pick 80GB when the weights plus KV cache or optimizer state pass 40 GB: QLoRA on a 70B model, 13B inference with long context, or larger training batches. For smaller models the 40GB card does the same work for less per hour.

NVIDIA · Ampere · 2020

NVIDIA A100 80GB. 80 GB HBM2e, 312 TFLOPS FP16 tensor.

Name: NVIDIA A100 80GB GPU Cloud Instance
Brand: NVIDIA
Price: 1.73 USD
Availability: InStock

Ampere with 80 GB of HBM2e, the value pick for 80 GB datacenter work.

Launch A100 80GB Get a reserved quote

A100 80GB on demand

from/hr per GPU

Availability: Available now · 6 regions
By region and size: $1.73 to $4.26/hr
Billing: By the second on most GPUs, a few by the minute
8+ GPUs or a month: Launch any size yourself, and prepay 1 to 12 months when you launch an instance. Custom quotes on request.

The machine

80 GB of HBM2e. At up to 2.0 TB/s.

Memory: 80 GB; HBM2e
Bandwidth: up to 2.0 TB/s; Memory to compute
FP16 tensor: 312; TFLOPS
Per node: 8 GPUs; up to 400 W each

Price across clouds

NVIDIA A100 80GB, cloud by cloud. Sorted by price.

On-demand rates per GPU-hour from each cloud's public price page on 22 Oct 2026: RunPod's Secure Cloud, and Lambda's lowest per-GPU rate across its cluster sizes. Runcrate's row is today's live starting rate.

Runcrate$1.73/hr

RunPod$1.79/hr

Lambda$2.79/hr

Workloads

What teams run on the A100 80GB. Each one opens the launch form, sized.

Llama 13B InferenceBF16 weights (26 GB) leave 54 GB for KV cacheLaunch it QLoRA Fine-tuneUp to 70B in 4-bit on 1 GPULaunch it Multi-GPU Training8x A100 80GB in one nodeLaunch it Diffusion GenerationSDXL at native 1024x1024Launch it

Launch it

Ask your agent for an A100 80GB. Or read the call it makes.

Claude Code or Codex runs the runcrate CLI for you: it starts the instance, copies your files over and runs the job. The SDKs make the same request.

Start an A100 80GB on Runcrate called my-gpu, copy train.py to it and run it.

what your agent runsbash

1runcrate instances create --name my-gpu --gpu A100_80G
2#   Instance queued: my-gpu (A100_80G)
3runcrate instances status my-gpu
4runcrate cp ./train.py my-gpu:/root/workspace/
5runcrate ssh my-gpu -- python /root/workspace/train.py

Spec sheet

How the A100 80GB compares. Against its closest alternatives.

FP16 tensor

NVIDIA A100 80GB312 TFLOPS

NVIDIA A100 40GB312 TFLOPS

NVIDIA L40S362 TFLOPS

AMD MI250X383 TFLOPS

Memory

NVIDIA A100 80GB80 GB

NVIDIA A100 40GB40 GB

NVIDIA L40S48 GB

AMD MI250X128 GB

Bandwidth

NVIDIA A100 80GBup to 2.0 TB/s

NVIDIA A100 40GB1.6 TB/s

NVIDIA L40S864 GB/s

AMD MI250X3.3 TB/s

Alternatives

Alternatives to the A100 80GB. Same account, different price and performance.

GPU	Memory · FP16 tensor	Stands out	Price gap	From
NVIDIA A100 80GBThis GPU	80 GB HBM2e · 312 TFLOPS			$1.73/hr
NVIDIA A100 40GB	40 GB HBM2 · 312 TFLOPS	-40 GB memory	+15%	$1.99/hr
NVIDIA L40S	48 GB GDDR6 · 362 TFLOPS	-32 GB memory	-34%	$1.14/hr
AMD MI250X	128 GB HBM2e · 383 TFLOPS	+48 GB memory		Not on demand
NVIDIA RTX 5090	32 GB GDDR7 · 209.5 TFLOPS	-48 GB memory		Not on demand

Full specs

NVIDIA A100 80GB specifications.

Memory

Memory: 80 GB
Type: HBM2e
Bandwidth: up to 2.0 TB/s

Compute

FP32: 19.5 TFLOPS
FP16 tensor: 312 TFLOPS
INT8: 624 TOPS
Tensor cores: 432
CUDA cores: 6,912

Power and form

TDP: up to 400 W
Architecture: Ampere
Released: 2020

Node

GPUs per node: 8
Maker: NVIDIA

Questions

Renting the A100 80GB. What buyers ask first.

Other GPUs

Browse every GPU by architecture.

NVIDIA Blackwell

NVIDIA Hopper

NVIDIA Ampere

NVIDIA Ada Lovelace

NVIDIA Volta

NVIDIA V100$3.04/hr

AMD Instinct