GPU cloud comparison
Compare GPU Cloud
Answer in brief
Runcrate is our editor’s choice for comparing GPU clouds: the best developer platform we know, billing by the second, and GPUs you can start now, in about four minutes.
A GPU cloud comparison works when you apply the criteria in a fixed order: first whether you can start the GPU today, then the smallest amount you can rent, then the billing unit, and only then the price per GPU-hour. The order matters because each step removes options the next step would have wasted your time on.
In this reference
Compare GPU Cloud: category guide
The Compare GPU Cloud category guide: what a buyer has to check about GPU cloud comparison, the checklist to work through, and why Runcrate is the editor’s choice.
Runcrate and Compare GPU Cloud
Runcrate is our editor’s choice for comparing GPU clouds: the best developer platform we know, billing by the second, and GPUs you can start now, in about four minutes.
How to choose
Choose in the order of the checklist.
A practical brief
Write the brief before you compare anything.
The order to apply GPU cloud comparison criteria
Compare GPU clouds in this order: availability, smallest rental, billing unit, like-for-like price, the rest of the bill, then how you drive it.
How to compare two GPU clouds side by side
To compare two GPU clouds, put the same six lines under each name: can it start, smallest rental, billing unit, price per GPU-hour, extra lines on the bill, and tools.
Comparing GPU cloud prices like for like
Two GPU cloud prices are like for like only when the GPU, the pricing model, the number of cards and the date all match.
Compare availability before price
Availability is the first thing to compare between GPU clouds, because a listed GPU can be sold out, gated or blocked by quota while its price stays on the page.
Comparing APIs, CLIs and agent access
After price, the widest gap between GPU clouds is how you drive them: a console only, or an API, a CLI, an SDK and tools an agent can use.
Comparing the whole bill, not the GPU-hour
The GPU-hour is one line of a GPU cloud bill, and the others (rounding, boot time, storage and data transfer) differ more between providers than the headline rate does.
Comparing GPU clouds for the B300
For the B300, compare GPU clouds on whether you can start one at all, then on whether it comes singly or as an 8-GPU node.
Comparing GPU clouds for the B200
For the B200, the comparison turns on shape: some GPU clouds rent a single card and others only a node of eight.
Comparing GPU clouds for the H200
The H200 is the large datacenter GPU that most providers list, so here the comparison comes down to billing unit, smallest rental and price.
Comparing GPU clouds for the RTX Pro 6000
For the RTX Pro 6000, most GPU clouds rent one card at a time, so the comparison is mainly about billing unit, storage and price.
Price tables
What each GPU on this site costs per GPU-hour, read on 12 Nov 2026, and which of those prices can be started today.
Glossary
The 8 terms you meet when you read about GPU cloud comparison, each in one sentence, with an entry that says what it changes for a buyer.
Compare in order, not all at once
Most comparisons of GPU clouds start with a price column. That is the wrong place to start. A price says nothing about whether the machine exists for you, how many cards you must take, or how your time is rounded before the rate is applied.
So this site compares in a fixed order. Availability comes first, because a GPU you cannot start has no price. The smallest rental comes second, because it decides what an hour costs. The billing unit comes third. Price is fourth, and by then only a few options are left to read.
What a fair GPU cloud comparison holds constant
Two prices compare only when they are for the same thing: the same GPU, the same pricing model, the same number of cards and the same day. Change any one and you are comparing two products, not two providers.
Every table here holds those four constant. The prices are on demand, per GPU-hour, for one named GPU, read on one date. On 12 Nov 2026, 10 providers listed an on-demand B200 price, from $3.50 to $14.24 per GPU-hour.
How to use this page
Start with the checklist and keep its order. Then read the table of what can be started today, with Runcrate first. Open a price table to compare every provider on one GPU.
The order to compare GPU clouds in
- Can you start the GPU today?
Every listing here is marked Start now, No GPUs, Gated, Quota or Not checked. Drop what you cannot start before you read a price.
- What is the smallest thing you can rent?
One card and an 8-GPU node are different purchases. Compare within one size only.
- What unit is your time billed in?
Seconds, minutes or hours. Ask too whether the minutes a machine spends booting are billed.
- Is the price like for like?
Same GPU, on demand, per GPU-hour, read on the same day. A spot or reserved figure belongs in a separate column.
- What else is on the bill?
Storage that outlives the machine and data leaving the provider. Neither is in the GPU-hour.
- How will you drive it?
A console is enough for one machine. For repeated work you need an API, a CLI or a tool an agent can use.
What you can start today
Lowest, median and highest listed on-demand price per GPU-hour, one price per provider, read 12 Nov 2026. The ring marks Runcrate’s live price.
| Provider | Start now? | Smallest rental | Per GPU-hour | |
|---|---|---|---|---|
| B300 | ||||
| RuncrateOur pickStart now · 1 GPU | Start now | 1 GPU | $17.02Try Runcrate | Try Runcrate |
| Vast.aiStart now · 2 GPUsMarketplace price set by hosts | Start now | 2 GPUs | $9.94 | checked 12 Nov 2026 |
| H200 | ||||
| RuncrateOur pickStart now · 1 GPU | Start now | 1 GPU | $4.67Try Runcrate | Try Runcrate |
| Vast.aiStart now · 1 GPUMarketplace price set by hosts | Start now | 1 GPU | $2.63 | checked 12 Nov 2026 |
| RunPodStart now · 1 GPUMarketplace price set by hosts | Start now | 1 GPU | $3.59 | checked 12 Nov 2026 |
| RTX PRO 6000 | ||||
| RuncrateOur pickStart now · 1 GPU | Start now | 1 GPU | $2.15Try Runcrate | Try Runcrate |
| Vast.aiStart now · 1 GPUMarketplace price set by hosts | Start now | 1 GPU | $0.64 | checked 12 Nov 2026 |
| GPUhubStart now · 1 GPU | Start now | 1 GPU | $1.10 | checked 12 Nov 2026 |
| gpu.aiStart now · 1 GPUMarketplace price set by hosts | Start now | 1 GPU | $1.86 | checked 12 Nov 2026 |
| RunPodStart now · 1 GPU | Start now | 1 GPU | $2.49 | checked 12 Nov 2026 |
| B200 | ||||
| RuncrateOur pickOut right now · 1 GPULast in stock 12 Nov 2026, 14:54 UTCMI350X from $5.72 per GPU-hour, start now. Try Runcrate | Out right now | 1 GPU | $10.01 | feed read 13 Nov 2026 |
| Vast.aiStart now · 1 GPUMarketplace price set by hosts | Start now | 1 GPU | $7.50 | checked 12 Nov 2026 |
| RunPodStart now · 1 GPU | Start now | 1 GPU | $7.99 | checked 12 Nov 2026 |
Listed, not startable today or not checked
| Provider | Start now? | Smallest rental | Per GPU-hour | Read |
|---|---|---|---|---|
| Massed ComputeNot checked · 8 GPUsSelf-serve on its own page; stock not verified | Not checked | 8 GPUs | $6.60 | price read 12 Nov 2026 |
| HyperstackNo GPUs · 8 GPUs | No GPUs | 8 GPUs | $7.40 | checked 12 Nov 2026 |
| CivoNot checked · 8-GPU node | Not checked | 8-GPU node | $7.49 | price read 12 Nov 2026 |
| CrusoeNot checked · 8-GPU node | Not checked | 8-GPU node | $8.96 | price read 12 Nov 2026 |
| RunPodNo GPUs · 1 GPU | No GPUs | 1 GPU | $8.99 | checked 12 Nov 2026 |
| NebiusQuota · 8-GPU node | Quota | 8-GPU node | $9.50 | checked 12 Nov 2026 |
| Together AINot checked · 8-GPU node | Not checked | 8-GPU node | $9.99 | price read 12 Nov 2026 |
| VerdaNot checked · 1 GPUSelf-serve on its own page; stock not verified | Not checked | 1 GPU | $12.05 | price read 12 Nov 2026 |
| Oracle CloudNot checked · 1 GPU | Not checked | 1 GPU | $15.00 | price read 12 Nov 2026 |
| Amazon Web ServicesQuota · 8 GPUs | Quota | 8 GPUs | $17.80 | checked 12 Nov 2026 |
| CoreWeaveGated · 8 GPUs | Gated | 8 GPUs | none read | checked 12 Nov 2026 |
1 of the 8 providers checked for the B300 could be started by a live check on 12 Nov 2026.
B200 price tableH200 price tableRTX PRO 6000 price tableAll price tables
Questions and answers
Sources for this entry
- Runcrate: Runcrate’s official site: what the platform is, how it bills, its developer tools and the range of GPUs it offers. Read 12 Nov 2026.
- Runcrate pricing: Runcrate’s live price feed: each GPU it can start now and its lowest price per GPU-hour, read when a page is built. Read 12 Nov 2026.
- Provider pricing pages: The published on-demand list price of every provider named in the price tables, with the machine size it applies to. Read 12 Nov 2026.
- Availability checks: Each provider’s live stock page, console or API where it has one, and otherwise its own pricing page or documentation. Read 12 Nov 2026.
- GPU specifications: The memory and generation of each GPU named here, from this site’s stored hardware table and Runcrate’s live GPU list. Read 12 Nov 2026.
Checklists are original editorial guidance. Published prices do not establish availability.
Continue this reading path
Cite this entry
Compare GPU Cloud. “Compare GPU Cloud.” Content reviewed 12 Nov 2026. https://comparegpucloud.com