NewEco lane for GPT-6 Astra: the same model at a lower price

Currency

You always top up in euro. Other currencies are shown at today's ECB rate. ECB reference rates

Sign inGet started
NexPodsComing soon

A GPU by the hour. Billed per minute.

Start a pod from the same account and the same credit as your API calls. A stopped pod only costs its storage.

  • From €0.26 an hourRTX 3090 and 4090 for development and inference.
  • Up to 45% below RunPodBetween 6% (A100 80GB) and 45% (RTX 4090) below RunPod's list prices.
  • Billed per minuteStop the pod and compute billing stops. Storage is billed per GB.
  • Same account and balanceYour API credit pays for pods. One invoice, one team.
Prices

Every GPU, per hour.

Prices per GPU hour in your currency, billed per minute, next to RunPod's list price where we are cheaper.

MemoryAll24 GB80 GB or more141 GB or more
GPUMemoryPer hourRunPodYou saveAvailabilityAccessLocation
H200141 GB€3.36€4.82−30%LimitedTier 1US
B200192 GB€5.28€7.28−28%LimitedTier 2US
GB200186 GB€8.38––On requestTier 2US

We show the RunPod list price only where we are cheaper, as checked on 15 November 2026. Access grows with your account: tier 0 starts up to 2 RTX pods at a time, tier 1 adds A100, H100 and H200 after 7 days and €100 paid, tier 2 is approved by hand. Locations and availability are examples.

GPU savings

What you save against RunPod

GPU
You save per month€547.05€6,565 a year · 25% less than RunPod

RunPod list prices as checked on 15 November 2026. NexPods prices are checked again before launch.

Get started
Templates

Start from a template.

vLLM

Serve an open model behind an API

ComfyUI

Image and video workflows

Jupyter

Notebooks with the GPU ready

Ubuntu + CUDA

A clean machine, your setup

CUDA 12

Drivers and CUDA ready for your own stack

Your own image

Bring your own container image

How pods work

Built for short jobs and long runs.

Per minute

Pay for the minutes a pod runs.

Billing starts when the pod is ready and stops when you stop it.

09:00H100start
09:01H1001 min€0.0454
11:47H100stop€7.5877
11:47volume100 GB
167 minutes of H100€7.59
One balance

Pods and API from one credit.

No second account and no second invoice.

Balance€1,987.60
Access

SSH and Jupyter.

Connect from your own terminal or open the notebook in the browser.

ssh root@pod-7f2a.nexpods.net
Storage

A volume that stays.

Your data stays on the volume while the pod is stopped.

Low balance

A warning before it stops.

With less than 15 minutes of credit left you get an email. At zero the pod stops.

Warning at15 min
Access tiers

Access grows with your account.

Tier 0RTX cards, up to 2 pods
Tier 1A100, H100 and H200
Tier 2B200, GB200 and more than 8 GPUs

Start a pod in five steps.

01Pick a GPUFrom RTX 4090 to B200.
02Pick a templatevLLM, ComfyUI, Jupyter or Ubuntu.
03Add storageKept while the pod is stopped.
04Add an SSH keyConnect from your own terminal.
05Check the costPer hour, before you start.
Questions

About pods.

How is a pod billed?

Per minute, from the moment the pod is ready until you stop it. A stopped pod only costs its storage, per GB.

What happens when my balance runs out?

With less than 15 minutes of credit left we email you. At zero the pod stops. After 7 days without a top-up the storage is removed, with emails on day 1, 3 and 6.

Which GPUs can I start?

New accounts start RTX cards, up to 2 at a time. After 7 days and €100 paid you can start A100, H100 and H200. B200 and GB200 are approved by hand.

Where do the pods run?

In data centers in the EU and the US. The location is shown per GPU before you start.

What is not allowed?

Cryptomining and anything else in the acceptable use policy. A pod that breaks the rules is stopped and the account is reviewed.