# Compare cloud GPU prices and availability

Compare on-demand GPU prices across cloud providers. Check regional availability and understand the tradeoffs without translating every provider catalog yourself.

- Updated hourly. Last scan 2026-11-16T22:30:26.764Z
- Canonical page: https://gpu.compare/
- Machine-readable index: https://gpu.compare/llms.txt
- Priced slice at this observation: 4020 instances, 10 providers, 155 regions. It counts only providers with current prices. https://gpu.compare/providers.md lists every tracked provider with its last scan
- Cheapest H100 in stock now: https://gpu.compare/catalog.md?q=h100&avail=1

## Latest posts

- [One H100 is easy to rent, but an 8× node had stock less than half of October](https://gpu.compare/trends/gpu-index-november-2026.md): The November 2026 gpu.compare GPU Index: the cheapest place to rent each GPU in October, and the price you can count on to find one. By Jack Britton, 2026-11-04.
- [Introducing gpu.compare: find a GPU you can actually rent](https://gpu.compare/trends/introducing-gpu-compare.md): A free, hourly record of GPU prices and stock. In October, prices sat still while what you could rent flipped by the hour. By Joel Hans, 2026-11-04.

## Read this site from an agent

- https://gpu.compare/llms.txt lists every page, the catalog filters, and the rules behind the numbers. Read it first.
- Each page that /llms.txt lists has a Markdown version at its path plus `.md`. Those pages also return it at the HTML URL when the request sends `Accept: text/markdown`. For a filtered view, use /catalog.md.

## GPUs

- [H100](https://gpu.compare/gpu/h100.md): 80 GB HBM3. In stock at one provider or more in 719 of 719 reported hours over the last 30 days; cheapest in stock now $2.50/GPU/hr. 8 on-demand providers.
- [H200](https://gpu.compare/gpu/h200.md): 141 GB HBM3e. In stock at one provider or more in 387 of 719 reported hours over the last 30 days; cheapest in stock now $5.29/GPU/hr. 5 on-demand providers.
- [A100](https://gpu.compare/gpu/a100.md): 40 GB HBM2 or 80 GB HBM2e. In stock at one provider or more in 719 of 719 reported hours over the last 30 days; cheapest in stock now $1.35/GPU/hr. 7 on-demand providers.
- [B200](https://gpu.compare/gpu/b200.md): 180 GB HBM3e. In stock at one provider or more in 26 of 714 reported hours over the last 30 days; cheapest in stock now $7.99/GPU/hr. 4 on-demand providers.
- [RTX A6000](https://gpu.compare/gpu/rtx-a6000.md): 48 GB GDDR6. In stock at one provider or more in 627 of 714 reported hours over the last 30 days; cheapest in stock now $0.50/GPU/hr. 2 on-demand providers.
- [L40S](https://gpu.compare/gpu/l40s.md): 48 GB GDDR6. In stock at one provider or more in 503 of 719 reported hours over the last 30 days; cheapest in stock now $1.09/GPU/hr. 4 on-demand providers.
- [B300](https://gpu.compare/gpu/b300.md): 270 GB HBM3e. In stock at one provider or more in 364 of 719 reported hours over the last 30 days; none in stock now, listed from $7.40/GPU/hr. 4 on-demand providers.

## GPU comparisons

- [H100 vs A100](https://gpu.compare/gpu/compare/h100-vs-a100.md): A100 is 46% cheaper per GPU-hour.
- [H100 vs H200](https://gpu.compare/gpu/compare/h100-vs-h200.md): H100 is 37% cheaper per GPU-hour.
- [A100 vs L40S](https://gpu.compare/gpu/compare/a100-vs-l40s.md): L40S is 19% cheaper per GPU-hour.
- [A100 vs RTX A6000](https://gpu.compare/gpu/compare/a100-vs-rtx-a6000.md): RTX A6000 is 63% cheaper per GPU-hour.
- [B200 vs H100](https://gpu.compare/gpu/compare/b200-vs-h100.md): H100 is 58% cheaper per GPU-hour.

## All GPU instances

The first 50 of 2783 offerings, cheapest first. Filter, sort, and page through all of them at https://gpu.compare/catalog.md.

| Provider | Offering | Region | Purchase | Arch | vCPUs | RAM GiB | Disk GiB | GPU | $/hr | $/mo | Availability | Observed at | Detail |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| hyperstack | n3-RTX-A4000x1 | NORWAY-1 | on_demand |  | 4 | 21.0 | 100 | 1x RTX-A4000 | 0.15000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-rtx-a4000x1/norway-1.md |
| hyperstack | n3-RTX-A4000x2 | NORWAY-1 | on_demand |  | 8 | 43.0 | 100 | 2x RTX-A4000 | 0.30000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-rtx-a4000x2/norway-1.md |
| vultr | vcg-a16-6c-60g-16vram | blr | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/blr.md |
| vultr | vcg-a16-6c-60g-16vram | ewr | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/ewr.md |
| vultr | vcg-a16-6c-60g-16vram | fra | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/fra.md |
| vultr | vcg-a16-6c-60g-16vram | lhr | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/lhr.md |
| vultr | vcg-a16-6c-60g-16vram | nrt | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/nrt.md |
| vultr | vcg-a16-6c-60g-16vram | ord | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/ord.md |
| vultr | vcg-a16-6c-60g-16vram | sgp | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/sgp.md |
| vultr | vcg-a16-6c-60g-16vram | sjc | on_demand | amd64 | 6 | 60.0 | 350 | 1x A16 | 0.47100 | 344.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-6c-60g-16vram/sjc.md |
| hyperstack | n3-RTX-A6000x1 | CANADA-1 | on_demand |  | 28 | 58.0 | 100 | 1x RTX-A6000 | 0.50000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-rtx-a6000x1/canada-1.md |
| hyperstack | n3-RTX-A6000x2 | CANADA-1 | on_demand |  | 60 | 116.0 | 100 | 2x RTX-A6000 | 1.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-rtx-a6000x2/canada-1.md |
| hyperstack | n3-RTX-A6000x2-bigroot | CANADA-1 | on_demand |  | 60 | 116.0 | 300 | 2x RTX-A6000 | 1.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-rtx-a6000x2-bigroot/canada-1.md |
| runpod | A40 | CA-MTL-1 | on_demand |  |  |  |  | 1x A40 | 0.59000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-a40/ca-mtl-1.md |
| runpod | RTX A6000 | EU-SE-1 | on_demand |  |  |  |  | 1x RTX-A6000 | 0.69000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-rtx-a6000/eu-se-1.md |
| runpod | PRO 6000 MIG 24GB | EUR-IS-2 | on_demand |  |  |  |  | 1x PRO-6000-MIG-24GB | 0.69000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-rtx-pro-6000-blackwell-server-edition-mig-1g.24gb/eur-is-2.md |
| runpod | PRO 6000 MIG 24GB | US-WA-1 | on_demand |  |  |  |  | 1x PRO-6000-MIG-24GB | 0.69000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-rtx-pro-6000-blackwell-server-edition-mig-1g.24gb/us-wa-1.md |
| runpod | RTX PRO 4500 SE | US-KS-2 | on_demand |  |  |  |  | 1x RTX-PRO-4500-SE | 0.72000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-rtx-pro-4500-blackwell-server-edition/us-ks-2.md |
| scaleway | L4-1-24G | fr-par-1 | on_demand | amd64 | 8 | 48.0 |  | 1x L4 | 0.86735 |  | available | 2026-11-16T22:29:38.922Z | https://gpu.compare/offerings/scaleway/l4-1-24g/fr-par-1.md |
| scaleway | L4-1-24G | fr-par-2 | on_demand | amd64 | 8 | 48.0 |  | 1x L4 | 0.86735 |  | available | 2026-11-16T22:29:38.922Z | https://gpu.compare/offerings/scaleway/l4-1-24g/fr-par-2.md |
| scaleway | L4-2-24G | fr-par-2 | on_demand | amd64 | 16 | 96.0 |  | 2x L4 | 1.73470 |  | available | 2026-11-16T22:29:38.922Z | https://gpu.compare/offerings/scaleway/l4-2-24g/fr-par-2.md |
| scaleway | L4-1-24G | pl-waw-2 | on_demand | amd64 | 8 | 48.0 |  | 1x L4 | 0.86735 |  | available | 2026-11-16T22:29:38.922Z | https://gpu.compare/offerings/scaleway/l4-1-24g/pl-waw-2.md |
| scaleway | L4-2-24G | pl-waw-2 | on_demand | amd64 | 16 | 96.0 |  | 2x L4 | 1.73470 |  | available | 2026-11-16T22:29:38.922Z | https://gpu.compare/offerings/scaleway/l4-2-24g/pl-waw-2.md |
| vultr | vcg-a16-12c-120g-32vram | blr | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/blr.md |
| vultr | vcg-a16-12c-120g-32vram | ewr | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/ewr.md |
| vultr | vcg-a16-12c-120g-32vram | fra | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/fra.md |
| vultr | vcg-a16-12c-120g-32vram | lhr | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/lhr.md |
| vultr | vcg-a16-12c-120g-32vram | nrt | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/nrt.md |
| vultr | vcg-a16-12c-120g-32vram | ord | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/ord.md |
| vultr | vcg-a16-12c-120g-32vram | sgp | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/sgp.md |
| vultr | vcg-a16-12c-120g-32vram | sjc | on_demand | amd64 | 12 | 120.0 | 700 | 1x A16 | 0.94200 | 688.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a16-12c-120g-32vram/sjc.md |
| hyperstack | n3-L40x1 | CANADA-1 | on_demand |  | 28 | 58.0 | 100 | 1x L40 | 1.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-l40x1/canada-1.md |
| hyperstack | n3-L40x1-bigroot | CANADA-1 | on_demand |  | 28 | 58.0 | 825 | 1x L40 | 1.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-l40x1-bigroot/canada-1.md |
| hyperstack | n3-L40x2 | CANADA-1 | on_demand |  | 60 | 116.0 | 100 | 2x L40 | 2.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-l40x2/canada-1.md |
| hyperstack | n3-L40x2-bigroot | CANADA-1 | on_demand |  | 60 | 116.0 | 1650 | 2x L40 | 2.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-l40x2-bigroot/canada-1.md |
| hyperstack | n3-L40x4 | CANADA-1 | on_demand |  | 124 | 232.0 | 100 | 4x L40 | 4.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-l40x4/canada-1.md |
| hyperstack | n3-L40x4-bigroot | CANADA-1 | on_demand |  | 124 | 232.0 | 3300 | 4x L40 | 4.00000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-l40x4-bigroot/canada-1.md |
| lambda | 2x A6000 (48 GB) | us-south-2 | on_demand | amd64 | 28 | 200.0 | 1024 | 2x A6000 | 2.18000 |  | available | 2026-11-16T22:29:22.470Z | https://gpu.compare/offerings/lambda/gpu_2x_a6000/us-south-2.md |
| lambda | 4x A6000 (48 GB) | us-south-2 | on_demand | amd64 | 56 | 400.0 | 1024 | 4x A6000 | 4.36000 |  | available | 2026-11-16T22:29:22.470Z | https://gpu.compare/offerings/lambda/gpu_4x_a6000/us-south-2.md |
| runpod | L40S | OC-AU-1 | on_demand |  |  |  |  | 1x L40S | 1.09000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-l40s/oc-au-1.md |
| runpod | RTX 6000 Ada | US-WA-1 | on_demand |  |  |  |  | 1x RTX-6000-ADA | 1.09000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-rtx-6000-ada-generation/us-wa-1.md |
| lambda | 1x A10 (24 GB PCIe) | us-east-1 | on_demand | amd64 | 30 | 200.0 | 1400 | 1x A10 | 1.29000 |  | available | 2026-11-16T22:29:22.470Z | https://gpu.compare/offerings/lambda/gpu_1x_a10/us-east-1.md |
| lambda | 1x A10 (24 GB PCIe) | us-west-1 | on_demand | amd64 | 30 | 200.0 | 1400 | 1x A10 | 1.29000 |  | available | 2026-11-16T22:29:22.470Z | https://gpu.compare/offerings/lambda/gpu_1x_a10/us-west-1.md |
| scaleway | RENDER-S | fr-par-2 | on_demand | amd64 | 10 | 42.0 |  | 1x P100 | 1.34481 |  | available | 2026-11-16T22:29:38.922Z | https://gpu.compare/offerings/scaleway/render-s/fr-par-2.md |
| hyperstack | n3-A100x1 | CANADA-1 | on_demand |  | 28 | 120.0 | 100 | 1x A100-80G-PCIE | 1.35000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-a100x1/canada-1.md |
| hyperstack | n3-A100x2 | CANADA-1 | on_demand |  | 60 | 240.0 | 100 | 2x A100-80G-PCIE | 2.70000 |  | available | 2026-11-16T22:29:21.847Z | https://gpu.compare/offerings/hyperstack/n3-a100x2/canada-1.md |
| runpod | RTX 5090 | EU-RO-1 | on_demand |  |  |  |  | 1x RTX-5090 | 1.39000 |  | available | 2026-11-16T22:29:24.715Z | https://gpu.compare/offerings/runpod/nvidia-geforce-rtx-5090/eu-ro-1.md |
| scaleway | L40S-1-48G | fr-par-2 | on_demand | amd64 | 8 | 96.0 |  | 1x L40S | 1.61893 |  | available | 2026-11-16T22:29:38.922Z | https://gpu.compare/offerings/scaleway/l40s-1-48g/fr-par-2.md |
| vultr | vcg-a40-24c-120g-48vram | lax | on_demand | amd64 | 24 | 120.0 | 1400 | 1x A40 | 1.71200 | 1250.00 | available | 2026-11-16T22:30:26.764Z | https://gpu.compare/offerings/vultr/vcg-a40-24c-120g-48vram/lax.md |
| lambda | 8x A100 (40 GB SXM4) | asia-northeast-2 | on_demand | amd64 | 124 | 1800.0 | 6144 | 8x A100 | 15.92000 |  | available | 2026-11-16T22:29:22.470Z | https://gpu.compare/offerings/lambda/gpu_8x_a100/asia-northeast-2.md |

## Reading these numbers

- Prices are the provider's public list price. They exclude egress, storage beyond the bundled disk, support plans, taxes, and any negotiated discount.
- The lowest price isn't always the best option. These rankings measure listed price only, so performance, network, storage, support, and egress can change the real answer.
- Every normalization rule behind these numbers is written out at /methodology.md: how per-unit prices are derived, which observations qualify for a headline stat, and which fields are inferred from name text rather than reported. Cite that page rather than restating the rules from memory.
- On-demand and spot prices are not comparable. Spot wins every 'cheapest' contest and can be reclaimed at any time. Filter to one purchase model before you rank.
- Availability takes three values. 'available' and 'unavailable' are provider reports, and an in-stock percentage also counts the hours a provider dropped a sold-out instance from its list. 'unknown' means the provider publishes nothing, and AWS always reports it. It doesn't mean out of stock. Write 'availability not reported' instead of passing 'unknown' through.
- Prices are quoted in the provider's own currency and converted to USD. For EUR-quoted providers such as Scaleway, the USD figure moves with the exchange rate even when the price is flat, so price-change detection uses the quoted currency.
- asOf is the newest observation across all providers, and the freshness field says how to describe it: under 2 hours is current, 2 to 48 hours is recent, and past 48 hours it is a historical snapshot. Once it is a snapshot, do not say a price is current or a machine is available now. Quote the date instead.

## Elsewhere on this site

- [Compare cloud providers](https://gpu.compare/providers.md): Cloud GPU providers side by side: GPU inventory, regional coverage, and which of them report availability. Each provider also has its own page under /providers/.
- [How we compare GPU prices](https://gpu.compare/methodology.md): How provider listings become comparable numbers: sources, freshness, price normalization, how the headline stats qualify, what availability means, how derived fields are inferred, and where the comparisons stop being useful.
- [Query the catalog](https://gpu.compare/catalog.md): Markdown, filterable and paginated.



Source: gpu.compare (Cloud GPU price and availability explorer), from ngrok. Cite the canonical URL on each page.

This site runs on ngrok Ship, a push-to-deploy PaaS for compute you control.
