# Rent the GPU for your next experiment.

Reference page: [IteraGPU — GPU rental for your ML experiments: 24 to 192 GB · IteraGPU](https://iteragpu.com/en/ressources)

Give your ML campaign the memory it needs, without buying a card. IteraGPU offers fifteen NVIDIA and AMD configurations, available to rent in crypto without KYC for 3, 7 or 30 days.

USD 23.57 for 1 batch of NVIDIA RTX A5000 24GB · 1 GPU with 24 GB · 3 days. Direct crypto payment, with no prior top-up.

- 24 to 192 GB per card
- NVIDIA or AMD
- Three durations to choose from

[Choose my GPU](https://iteragpu.com/en/ressources#catalogue) · [Configure NVIDIA RTX A5000 · 3 days](https://iteragpu.com/en/checkout?profile=rtx-a5000&days=3)

Account required: first name, last name, email and password. No KYC procedure and no ID document.

## Your rental in plain terms

**IteraGPU offer**

| Reference point | Condition |
| --- | --- |
| Hardware | 15 NVIDIA and AMD configurations; 24 to 192 GB per card. |
| Rental unit | 1 lot = 1 GPU, except B200: 2 GPUs per lot. From 1 to 10 lots per order. |
| Periods | Full packages of 3, 7 or 30 days; no automatic renewal. |
| Minimum to get started | USD 23.57 for 1 batch of NVIDIA RTX A5000 24GB, 1 GPU, 3 days; direct payment with no prior top-up. |
| Crypto payment | BTC of Bitcoin ; ETH of Ethereum ; USDT of Ethereum ; USDC of Ethereum ; SOL of Solana ; TRX of Tron ; USDT of Tron ; LTC of Litecoin. USD balance usable after confirmed top-up. |
| Account without KYC | First name, last name, email and password required. No KYC process and no ID document. |

## Three campaign budgets

### 3 days for your pilot

NVIDIA RTX A5000 24GB · 1 batch(es) · 1 GPU · 3 days · USD 23.57 in total. Lot price: USD 23.57.

Spend three days on a first trial: baseline run, memory measurement and saving results, for a workload suited to this card. 1 lot to configure.

[View configuration](https://iteragpu.com/en/ressources/rtx-a5000) · [Configure NVIDIA RTX A5000 · 3 days](https://iteragpu.com/en/checkout?profile=rtx-a5000&days=3)

### 7 days to compare two variants

NVIDIA A40 48GB · 2 batch(es) · 2 GPU · 7 days · USD 136.00 in total. Lot price: USD 68.00.

Assign a 48 GB card to each variant. Work on the same corpus and the same criteria, with a shared budget for both experiments. Choose 2 lots in the configurator.

[View configuration](https://iteragpu.com/en/ressources/a40) · [Configure NVIDIA A40 · 7 days](https://iteragpu.com/en/checkout?profile=a40&days=7)

### 30 days to track your campaign

NVIDIA A100 SXM 80GB · 1 batch(es) · 1 GPU · 30 days · USD 990.00 in total. Lot price: USD 990.00.

Keep an 80 GB card for thirty days to organize your test series, resume your experiments and gather their results. 1 lot to configure.

[View configuration](https://iteragpu.com/en/ressources/a100-sxm) · [Configure NVIDIA A100 SXM · 30 days](https://iteragpu.com/en/checkout?profile=a100-sxm&days=30)

## All plans

**IteraGPU prices in USD — full lot price for each period**

| Configuration | Memory per GPU | GPU / lot | Available lots | 3 days · USD | 7 days · USD | 30 days · USD |
| --- | --- | --- | --- | --- | --- | --- |
| NVIDIA RTX A5000 24GB | 24 Go | 1 | 51 | USD 23.57 | USD 55.00 | USD 190.00 |
| NVIDIA GeForce RTX 3090 24GB | 24 Go | 1 | 186 | USD 27.86 | USD 65.00 | USD 230.00 |
| NVIDIA A40 48GB | 48 Go | 1 | 37 | USD 29.14 | USD 68.00 | USD 240.00 |
| NVIDIA L4 24GB | 24 Go | 1 | 104 | USD 30.00 | USD 70.00 | USD 250.00 |
| NVIDIA RTX A6000 48GB | 48 Go | 1 | 65 | USD 35.14 | USD 82.00 | USD 290.00 |
| NVIDIA GeForce RTX 4090 24GB | 24 Go | 1 | 265 | USD 47.14 | USD 110.00 | USD 390.00 |
| NVIDIA RTX 6000 Ada 48GB | 48 Go | 1 | 26 | USD 55.71 | USD 130.00 | USD 460.00 |
| NVIDIA L40S 48GB | 48 Go | 1 | 74 | USD 63.43 | USD 148.00 | USD 520.00 |
| NVIDIA GeForce RTX 5090 32GB | 32 Go | 1 | 100 | USD 66.43 | USD 155.00 | USD 540.00 |
| NVIDIA A100 SXM 80GB | 80 Go | 1 | 78 | USD 120.00 | USD 280.00 | USD 990.00 |
| NVIDIA H100 PCIe 80GB | 80 Go | 1 | 27 | USD 180.00 | USD 420.00 | USD 1,490.00 |
| AMD Instinct MI300X 192GB | 192 Go | 1 | 8 | USD 192.86 | USD 450.00 | USD 1,590.00 |
| NVIDIA H100 SXM 80GB | 80 Go | 1 | 63 | USD 203.57 | USD 475.00 | USD 1,690.00 |
| NVIDIA H200 SXM 141GB | 141 Go | 1 | 19 | USD 274.29 | USD 640.00 | USD 2,290.00 |
| 2 × NVIDIA B200 SXM | 180 Go | 2 | 4 | USD 887.57 | USD 2,071.00 | USD 7,391.00 |

A B200 lot already includes its two cards. The prices in the table cover the entire period shown.

## Capacity for your campaign, without buying a card.

Mobilize your project's GPUs for 3, 7 or 30 days. You choose a configuration, a period and a number of lots: each listing brings together the memory per card, the available stock and the package price so you can prepare your purchase.

One lot contains one card for fourteen configurations and two cards for B200. This unit lets you compare an experiment on one GPU and a campaign spread across several lots, with a clearly defined total.

**The terms common to all IteraGPU rentals**

| Criterion | IteraGPU offer |
| --- | --- |
| Hardware choice | 15 NVIDIA and AMD configurations; from 24 to 192 GB nominal per card |
| Durations | 3, 7 or 30 days; minimum rental: 3 days |
| Quantity | From 1 to 10 lots in the configurator; availability shown for each offer |
| Lot composition | 1 GPU, except B200: 2 GPUs of 180 GB each |
| Price | From USD 23.57 for 1 lot of NVIDIA RTX A5000 24GB, i.e. 1 GPU of 24 GB, for 3 days. Price in USD per lot for the entire period; total = package × number of lots. |
| Payment | Crypto directly on the order, or available USD balance |
| Account | First name, last name, email and password; no ID document or KYC procedure required |

Check: 

## Find the cards worth choosing for your model.

An inference validation, a model adaptation and a series of training runs do not require the same memory. Start from your workload and your libraries, then compare the packages of the cards that meet those needs. IteraGPU gives you several options at every stage of the campaign.

- Start with 24 GB per card: the NVIDIA RTX A5000 24GB gives access to the entry price. The RTX 3090, RTX 4090 and L4 also offer 24 GB of capacity, with their own packages.
- Move up to 48 GB on a single card: compare A40, RTX A6000, RTX 6000 Ada and L40S to accommodate a workload that exceeds the capacity of a 24 GB card.
- Plan for at least 80 GB per card: look at A100 SXM 80 GB, H100 PCIe or SXM 80 GB, then H200 SXM 141 GB depending on your environment and the memory required.
- Choose a large capacity or several GPUs: MI300X offers 192 GB on an AMD card; one B200 lot brings together two NVIDIA cards of 180 GB each. Go with the organization your software can make use of.

Check: [NVIDIA RTX A5000 24GB · 24 GB](https://iteragpu.com/en/ressources/rtx-a5000) · [A40 · 48 GB](https://iteragpu.com/en/ressources/a40) · [A100 SXM · 80 GB](https://iteragpu.com/en/ressources/a100-sxm) · [MI300X · 192 GB](https://iteragpu.com/en/ressources/mi300x) · [B200 · two 180 GB cards](https://iteragpu.com/en/ressources/b200)

## A duration for every stage of your research.

Three days for a targeted validation; seven days to compare a series of variants; thirty days to organize a campaign with several stages. Choose the window that matches your work plan and see the package total before ordering.

For independent experiments, you can plan a single lot used in succession or several lots for parallel runs. Two standard lots represent two GPUs; two B200 lots represent four. The two B200 cards are already included in the price of each lot.

Check: [Compare campaign budgets](https://iteragpu.com/en/tarifs)

## Stay in control of your research, specify your needs.

You choose your models, your code and your dependencies. To select an offer, rely on the specifications of the listing and the requirements of your campaign.

If your decision depends on Jupyter access, file persistence, a preinstalled software version, RAM, storage, GPU interconnect, or provisioning time, have these points confirmed before ordering. Memory remains specific to each card: your software must organize the work across multiple GPUs.

Plans buy a period of capacity; throughput and the number of completed experiments depend on your workload. If your project requires hourly billing or a performance commitment, specify this condition before choosing.

Check: [Rental terms](https://iteragpu.com/en/conditions) · [Payment assets and networks](https://iteragpu.com/en/paiement-crypto) · [Specify a requirement before purchase](https://iteragpu.com/en/support)

## Prepare your access and environment

The stock shown counts available lots; it does not reserve a future date. Payment confirmation and provisioning are two distinct steps.

The amount shown corresponds to the package multiplied by the lots. No automatic surcharge is added by this calculation; this rule does not specify a tax-inclusive or tax-exclusive regime, nor any possible wallet fees.

The GPU capacity is stated in the offer. Software environment, RAM, storage, interconnect or lead time required for your campaign must be confirmed when they are decisive.

[Ordering terms](https://iteragpu.com/en/conditions) · [Sizing and selection criteria](https://iteragpu.com/en/dimensionner)

## Before ordering

### What is the initial amount to budget for renting an IteraGPU GPU?

IteraGPU's entry price is USD 23.57 for a NVIDIA RTX A5000 24GB lot of 24 GB for 3 days. This lot contains one card and the amount covers the entire period. The order can be paid directly in crypto, with no prior balance top-up. Any transfer fees should be checked in your wallet. The process involves no KYC procedure or identity document; an account with first name, last name, email and password is required.

### Which offers should I compare if my experiment requires 48 GB on a single card?

IteraGPU offers four 48 GB per card configurations: NVIDIA A40, RTX A6000, RTX 6000 Ada and L40S. Each one includes one GPU per lot. Compare their plans of the same duration and check your software's requirements. Identical memory capacity does not guarantee the same throughput or compatibility with every library.

### Do MI300X and B200 represent the same capacity for my model?

At IteraGPU, the AMD MI300X lot includes one 192 GB card; the NVIDIA B200 lot includes two 180 GB cards each. The two B200 memories do not automatically form a single unified space. The choice therefore depends on how your software partitions the work and its compatibility with the hardware family, in addition to price.

### Can I rent multiple GPUs to compare ML variants in parallel?

The IteraGPU configurator lets you choose from 1 to 10 lots, with availability shown per offer. One lot represents one card, except B200, which includes two. Multiple lots can correspond to independent experiments that you organize; ordering them does not prove an interconnect suited to distributed training. Have this requirement confirmed if it determines your purchase.

### Do I need an account and an identity document to order?

Ordering on IteraGPU uses an account with first name, last name, email and password. No identity document or KYC procedure is required. You can pay for the rental in crypto or use a sufficient USD balance. Contact details are still needed for tracking; paying without KYC does not mean purchasing without order data.

## Your next experiment has its configuration.

Choose the memory and duration suited to your workload, then configure your rental.

USD 23.57 for 1 batch of NVIDIA RTX A5000 24GB · 1 GPU with 24 GB · 3 days.

[Packages and scenarios](https://iteragpu.com/en/tarifs) · [Crypto payment](https://iteragpu.com/en/paiement-crypto) · [Ordering terms](https://iteragpu.com/en/conditions)

[Configure NVIDIA RTX A5000 · 3 days](https://iteragpu.com/en/checkout?profile=rtx-a5000&days=3)