Quick answer
What do the GPUwithCrypto terms say in brief?
You top up a prepaid balance in crypto, at least $25 at a time, you buy blocks of GPU-hours from it, and GPUwithCrypto charges those hours per minute while your GPU instances run. Instances are on-demand and stop when the GPU-hours of their GPU run out; their disk is wiped when you delete them. The acceptable use policy is part of these terms. The seller is SerpGoat LLC. No KYC is required.
1. Who agrees to these terms?
The account holder, who may be a person or an organisation, and SerpGoat LLC, which operates GPUwithCrypto. Opening an account, sending a top-up or starting an instance means accepting these terms and the acceptable use policy.
You must be legally able to enter into a contract where you live. If you act for an organisation, you confirm that you may bind it.
2. What do we sell?
Time on cloud GPU instances: NVIDIA H100 80GB, H200 141GB, B200, A100 80GB, L40S 48GB, GeForce RTX 4090 24GB and RTX 5090 32GB, rented by the minute. Each instance gives you whole GPUs, never a time-shared slice or a MIG partition. You get root SSH access to Ubuntu with the NVIDIA driver, CUDA and PyTorch installed, or start from your own Docker image.
Instances have 1 to 8 GPUs depending on the host. Before you start, the dashboard shows each host’s country, GPU count, form factor (SXM or PCIe) and interconnect. The GPU model, its hourly price and the disk size shown before you start are what the instance provides.
3. How does the account work?
Sign in with an account number; an email address is optional. Your account number is how you sign in, so store it carefully. With an optional email on file you can recover it; with no email, a lost number is gone for good.
Anything done with your account number, your API keys or the SSH keys you added is treated as done by you, so store them like passwords and replace a leaked key at once. Nobody from our team will ever ask for your account number, your SSH private key or your API keys, by email or in a chat.
4. How do I pay?
Prices are in US dollars. You fund a prepaid balance in BTC, USDT, USDC, ETH, SOL, LTC, at least $25 each time. Send the exact amount on the network the payment page names; the balance is credited once the transfer has the required confirmations on the blockchain. Network fees are paid by the sender.
Credit on your balance does not expire and buys blocks of GPU-hours for any GPU. When a top-up is only made to reach the minimum, the unspent part comes back on request within 30 days of payment, in USDT or USDC. The rest of a balance is not paid out unless the law requires it.
5. How is GPU time billed?
Instances are billed per minute at the hourly rate while they run, from the GPU-hours you bought for that GPU. Stopping an instance stops the charge at once. The pricing page lists the hourly rate of every GPU.
GPU-hours are sold in blocks of 10, 24, 168 or 720 hours for one GPU model, at the same hourly rate; instances of that GPU draw from them per minute, and unused hours never expire. A block is bought from the balance at the price shown when you buy it, and only instances of that GPU model draw from it.
6. How do on-demand instances run?
Once your payment is confirmed, you launch an instance from the dashboard; our team prepares it at the GPU provider and its SSH details appear there as soon as it runs. Billing starts only then. Instances are on-demand, not spot: nobody can take your GPU back while your balance covers it, and the instance runs until you stop it.
You start, stop and delete instances from the dashboard or the API. You alone decide what runs on them, within the acceptable use policy.
7. What happens when my GPU-hours run out?
When the GPU-hours of a GPU run out, its instances stop and their disks are kept; a new block of hours lets you start them again. A job that is still running at that moment is interrupted. Keep enough GPU-hours for the whole run, and save checkpoints regularly; a stop caused by used-up GPU-hours is not a fault of the service.
8. What happens to the data on an instance's disk?
Every instance has local NVMe disk, sized for its GPU and shown before you start. The disk is wiped when you delete the instance, so copy your checkpoints out first. A deleted disk cannot be recovered by anyone, including us.
We do not read the files on your disk, and we keep no backup of it. Copying your data in and out, and keeping your own copies, is your responsibility.
9. Is GPU availability guaranteed?
No. Stock changes through the day: the dashboard shows which GPUs are free right now, and nothing is charged until an instance starts. A GPU model can be fully in use when you want it, and buying a block of hours does not reserve a machine.
We do not promise an uptime figure. If an instance fails to start, or its GPU does not work and support cannot fix it, the refund rule in section 10 applies.
10. When do I get hours or money back?
If an instance fails to start, or its GPU does not work and support cannot fix it, the hours it used go back to your balance. Unused GPU-hours and balance never expire. Top-ups are not paid out unless the law requires it; the part added only to reach the minimum top-up comes back on request within 30 days of payment, in USDT or USDC, as long as it is unspent. The refund policy applies these rules to worked cases.
11. What rights do consumers keep?
If you are a consumer, the law of the country where you live may give you rights that a contract cannot remove, such as a right to withdraw from a purchase or a remedy for a faulty service. Those rights apply on top of these terms, and nothing in these terms reduces them.
12. Which rules of use apply?
The acceptable use policy is part of these terms. In short: No cryptocurrency mining, no password or hash cracking, no attacks on other systems (DDoS, scanning, exploitation), no malware, no spam, no child sexual abuse material, no intimate or deceptive images of real people without their consent, no weapons development and no work for sanctioned people or companies.
The licences of the models, datasets, images and software you run apply to you, and complying with them is your responsibility. You may build and sell your own products on your instances. Reselling raw access to our GPUs, as GPU rental, needs our written agreement.
13. Where is the service not available?
Not available to people or companies in Cuba, Iran, North Korea, Syria, Russia, Belarus, China, Yemen or Western Sahara, or in the occupied regions of Ukraine. Incoming payments are checked against public sanctions lists of blockchain addresses; a payment from a listed address is not credited.
By using the service you confirm that you are not located in those places, that you are not on a sanctions or export-control restricted-party list, and that you do not act for anyone who is. Accounts or payments linked to those places or parties are refused or closed, and the law may require us to hold their funds.
14. When can you suspend or close an account?
When an account breaks these terms or the acceptable use policy, when a payment turns out to be fraudulent or comes from a listed address, or when a law requires it. Reports go to [email protected]. An instance that breaks the acceptable use policy is stopped and the hours it used are not refunded; child sexual abuse material is reported to the authorities.
We tell you the reason unless the law forbids it. You can close your account at any time by writing to [email protected]; the privacy policy explains what is erased and what accounting law makes us keep.
15. What do you promise, and what not?
We provide the service as it stands on the day you use it: the GPU model, the disk and the software image shown before you start. We do not promise that a given model, framework or workload fits on a GPU or runs at a given speed, that the service is never interrupted, or that a use you plan is legal where you are. That is for you to check.
16. How far does your liability go?
As far as the law allows, our total liability for all claims is limited to the amount you paid us for the service the claim is about. We are not liable for indirect losses such as lost profit, lost data, a lost training run or a missed deadline. Nothing here limits a liability that the law does not allow us to limit.
17. Will prices or these terms change?
Yes, and every change carries a date. Our prices follow the rule on the pricing page and show the date of the price check they come from. A new price applies to GPU time used after it is published. GPU-hours already bought in a block stay hours: a later price change does not reduce them.
The date at the top of this page is the date of the version in force. A change applies from the day it is published here; use before that date stays under the earlier version.
18. Which law applies, and how do I reach you?
These terms are governed by the law of SerpGoat LLC's jurisdiction, except where the law of your country gives you protections that cannot be set aside.
Questions about these terms go to [email protected]. Abuse reports go to [email protected].