DEV Community: FastGPU

The big clouds charge a median 3.2x the cheapest GPU rate: 18 GPUs compared

FastGPU — Thu, 29 Oct 2026 16:24:53 +0000

Rent an H100 on Google Cloud and the cheapest on-demand rate is $5.38 per GPU-hour, sold as an 8-GPU machine. Rent the same card from Lium, a GPU marketplace, and it is $1.80. Same chip, same hour, three times the price.

That gap is not one odd card. We compared every GPU that both a hyperscaler (AWS, Google Cloud, Azure, Oracle Cloud) and at least one other cloud rent on demand, from the live prices FastGPU collects across the market. On October 29, 2026 at 16:20 UTC there were 18 such cards, and the cheapest big-cloud rate for a card was a median of 3.2x the cheapest rate at which it could be rented anywhere.

Card by card

The cheapest hyperscaler rate for each card against the market floor (the cheapest on-demand or community price at which the card could be rented that morning, in stock wherever any offer was), per GPU-hour:

GPU	Cheapest hyperscaler	Market floor	Hyperscaler / floor
V100	$2.49, Google Cloud (8-GPU instance)	$0.07, Clore.ai	35.6x
L4	$0.71, Google Cloud	$0.09, Vast.ai	7.9x
RTX PRO 4500	$2.52, AWS	$0.35, Clore.ai	7.2x
A100 40GB	$2.16, Google Cloud (16-GPU instance)	$0.35, Vast.ai	6.2x
L40S	$1.86, AWS	$0.34, Vast.ai	5.5x
A100 80GB	$3.22, Google Cloud (8-GPU instance)	$0.60, Vast.ai	5.4x
A10	$1.01, AWS	$0.24, Vast.ai	4.2x
MI300X	$6.00, Azure (8-GPU instance)	$1.71, TensorWave	3.5x
RTX PRO 6000	$3.36, AWS	$1.00, Vast.ai	3.4x
MI355X	$8.60, Oracle Cloud (8-GPU instance)	$2.85, TensorWave	3.0x
T4	$0.36, Google Cloud (4-GPU instance)	$0.12, Vast.ai	3.0x
H100 SXM	$5.38, Google Cloud (8-GPU instance)	$1.80, Lium	3.0x
B200	$14.00, Oracle Cloud (8-GPU instance)	$4.89, DeepInfra	2.9x
H100 NVL	$6.98, Azure	$2.59, RunPod	2.7x
H200	$7.91, AWS (8-GPU instance)	$3.00, Vast.ai	2.6x
B300	$15.00, Oracle Cloud (8-GPU instance)	$5.89, DeepInfra	2.5x
GB200	$16.00, Oracle Cloud (4-GPU instance)	$8.00, GMI Cloud	2.0x
GB300	$18.00, Oracle Cloud (4-GPU instance)	$12.93, Verda	1.4x

Each ratio is the first price divided by the second. The live table recomputes all of it on every visit and links every rate to the page that proves it.

Two patterns stand out. The gap is widest on older and mid-range cards (V100, L4, A100, L40S), where marketplaces full of hosts with paid-off hardware set the floor. And it narrows on the newest silicon: a GB300 costs only 1.4x more at Oracle than at the cheapest GPU cloud, because few providers rent one at all and none of them is cheap.

Cloud by cloud

Each big cloud's own on-demand rate against the floor, across every card it lists, gives a median per cloud (on the same snapshot):

AWS: 4.3x across 12 cards
Google Cloud: 5.4x across 7 cards
Azure: 5.5x across 11 cards
Oracle Cloud: 4.0x across 14 cards

A cloud with few cards is a small sample, so read these as a shape rather than a ranking. Each link goes to that cloud's full pricing guide, where every rate is a live figure.

What the premium buys, and what the floor leaves out

The comparison is honest only if you know what each side includes.

The floor is often a marketplace. Vast.ai, Clore.ai and similar platforms rent machines that independent hosts own. Network quality, data security and uptime vary by host, and none of them publishes an uptime SLA for a single machine. Across the last 25 daily fixings, Vast.ai alone held the cheapest price on 41% of GPU-days, and the cheapest provider for a card changed about one day in six.

Many hyperscaler rates come only as whole machines. The per-GPU rate on an 8-GPU instance is real, but the smallest bill is eight times it.

The big clouds charge for data leaving them. Moving 1,000 GB out costs $90.00 at AWS and $87.00 at Azure, (each after the first 100 GB a month), and $120.00 per 1,000 GiB at Google Cloud. Many GPU clouds charge nothing for egress. For a job that downloads a fine-tuned model or serves files, that can outweigh the GPU bill.

Meters differ. AWS and Google Cloud bill by the second, Azure by the minute; some GPU clouds bill by the hour, and several keep billing a stopped machine's disks. The hidden costs page lists each provider's egress, meter and uptime terms, each linked to the provider's own page.

What the premium does buy: region coverage, existing enterprise contracts and compliance, the surrounding platform (storage, networking, managed services), and published SLAs. For a team already committed to one of them, the premium is often the price of not moving data and not adding a vendor. For a portable training or inference job, it is the first number worth questioning.

How the numbers are made

FastGPU collects posted on-demand prices from each provider's own API, rate card or price list, every day and again while pages are being read, and normalises them to a per-GPU-hour rate. Spot capacity is excluded from both sides. A machine its provider no longer rents to new customers is left out, even when a price list still shows it. The methodology has the details, and the recorded prices are an open dataset under CC BY 4.0.

These figures are a snapshot from October 29, 2026. The live comparison shows today's.

15 GPU list prices rose in September and none fell. Today AWS raises Capacity Blocks 15%

FastGPU — Wed, 07 Oct 2026 01:52:01 +0000

In September 2026 we compared 176 posted on-demand GPU prices, each cloud's own published rate for each GPU it rents, between the first three and the last three days of the month. 15 ended the month higher, none ended it lower, and the other 161 did not move.

Today, October 7, AWS adds an increase of its own to EC2 Capacity Blocks for ML.

Everything below comes from the daily prices FastGPU records across GPU clouds, published as an open dataset, and from AWS's own pricing page. Every figure links to the page it comes from.

15 increases, no cuts

All 15 came from two providers: Verda (10) and RunPod (5). Marketplaces, where each host sets its own price, are not part of this count.

Provider	GPU	Sep 1 to 3	Sep 28 to 30	Change
Verda	H200	$4.00	$4.64	+16.0%
RunPod	A100 80GB	$1.39	$1.59	+14.4%
Verda	GB300	$8.62	$9.72	+12.8%
Verda	B300	$7.50	$8.45	+12.7%
Verda	L40S	$1.37	$1.54	+12.4%
Verda	B200	$6.11	$6.85	+12.1%
Verda	V100	$0.17	$0.19	+11.8%
Verda	RTX 6000 Ada	$1.04	$1.16	+11.5%
RunPod	A40	$0.44	$0.49	+11.4%
RunPod	RTX PRO 6000	$1.89	$2.09	+10.6%
RunPod	L40S	$0.99	$1.09	+10.1%
Verda	H100 SXM	$3.25	$3.56	+9.5%
RunPod	H100 SXM	$3.29	$3.49	+6.1%
Verda	A6000	$0.61	$0.64	+4.9%
Verda	RTX PRO 6000	$1.89	$1.96	+3.7%

The typical price rose less

The median of each provider's lowest posted on-demand price, counting only the providers that posted a price on all six of those days, so a provider joining or leaving never reads as a price move:

GPU	Sep 1 to 3	Sep 28 to 30	Change	Providers
H100 SXM	$3.55	$3.70	+4.3%	16
H200	$4.14	$4.38	+5.7%	14
B200	$6.18	$6.25	+1.1%	12
A100 80GB	$1.79	$1.79	-0.1%	15

The cheapest offer moves both ways

The floor, the lowest posted on-demand price outside the hyperscalers (geometric mean of the first three days against the last three), is set by one or two providers at a time, so it swings:

H100 SXM: $1.34 to $1.57 (+17.4%, set by 2 providers)
H200: $2.60 to $2.16 (-17.1%, set by 2 providers)
B200: $3.69 to $3.69 (0.0%, set by 1 provider)
A100 80GB: $0.64 to $0.53 (-16.7%, set by 1 provider)

So the list prices at the larger GPU clouds drifted up while the cheapest deal stayed a moving target. For a job of a few hours, the floor is what you pay; for a quarter's budget, the list-price trend matters more.

Same card, 44.0x apart

On September 30 a V100 cost from $0.07/hr at Vast.ai to $2.95/hr at Oracle Cloud across 8 providers: $2.95 / $0.067 (the low before rounding) = 44.0x. It was the widest spread of the month: the same card, 8 sellers.

AWS Capacity Blocks: 15% more from today, up to 26.5% outside the US

AWS's Capacity Blocks pricing page says:

Effective October 7, 2026, hourly rates per accelerator across all available AWS Regions will be: P6-B300 at $16.146 ($16.819 in AWS GovCloud Regions), P6-B200 at $14.208 ($14.801 in AWS GovCloud Regions), P5en at $7.895, P5e at $6.866, P5 at $5.970, P4de at $2.546, and P4d at $1.696.

Against the rates AWS published before, as recorded in our price-change log, per GPU per hour:

Instance	GPU	Before	From Oct 7	Change
P6-B300	B300	$14.040	$16.146	+15.0%
P6-B200	B200	$12.355	$14.208	+15.0%
P5en, US Regions	H200	$6.865	$7.895	+15.0%
P5en, Europe and Asia Pacific	H200	$6.241	$7.895	+26.5%
P5e	H200	$5.970	$6.866	+15.0%
P5, US Regions	H100 SXM	$5.191	$5.970	+15.0%
P5, Europe, Asia Pacific and South America	H100 SXM	$4.720	$5.970	+26.5%
P4de	A100 80GB	$2.214	$2.546	+15.0%
P4d	A100 40GB	$1.475	$1.696	+15.0%

AWS now charges one rate in every Region, so blocks that used to be cheaper outside the US (P5 and P5en) rise the most. AWS publishes these rates to a tenth of a cent; our log shows them rounded to the cent. AWS also says Capacity Block reservation prices are dynamic and are updated periodically based on supply and demand, and: "All other EC2 Capacity Block prices, as well as prices on fixed pricing models like On-Demand and Savings Plans, remain unchanged."

What it means if you rent

A Capacity Block is reserved capacity: you book a start date and a length, and AWS guarantees the GPUs for it. That is a different product from an on-demand GPU you can stop at any time. Still, for scale: the cheapest posted H100 SXM is $1.79/hr at Vast.ai as of 2026-10-07 01:46 UTC, the September-end median across 16 providers was $3.70/hr, and a P5 Capacity Block is now $5.970/hr.
If you were planning a reserved block, re-run the numbers: the reserve-or-rent calculator has the October 7 rates, and the GPU cluster page prices N GPUs for M months across providers' published term rates.
List prices change without notice. A price alert emails you when a GPU drops below a price you choose, with no account.

Data and method

Posted prices, not transactions. The September report is computed from the recorded daily prices once the month is over and then frozen; every dated change is in the price-change log, also as JSON at https://fastgpu.co/api/v1/price-changes. The daily series is an open dataset (CC BY 4.0). Live prices in this post are as of 2026-10-07 01:46 UTC; the linked pages show the current ones. Some provider links on FastGPU are affiliate links; tables and every "cheapest" figure are ordered strictly by price.

What it costs to rent an H100, B200 or RTX 4090 in September 2026: live prices from 28 GPU clouds

FastGPU — Sun, 27 Sep 2026 04:11:49 +0000

GPU rental prices have not converged. The same card can cost several times more depending on which cloud you rent it from, and the cheapest option changes through the day. We run FastGPU, which pulls prices from 28 GPU clouds (marketplaces, neoclouds, serverless platforms and the hyperscalers) and ranks them cheapest first.

Here is the market as it stood on September 27, 2026 at 04:11 UTC, and the few patterns that decide whether you overpay. Prices move daily, so every figure links to its live page.

The snapshot

Cheapest on-demand price per GPU-hour for each card, next to the cheapest on-demand price for the same card at a hyperscaler (AWS, Azure, Google Cloud or Oracle). Every figure in a row comes from that card's live page.

GPU	VRAM	Cheapest on-demand	Offered by	Cheapest hyperscaler	Gap (hyperscaler / cheapest)	Providers
H100 SXM	80 GB	$1.79/hr	Cudo Compute	$5.38/hr (Google Cloud)	5.38 / 1.79 = 3.0x	22
H200	141 GB	$2.60/hr	GMI Cloud	$7.91/hr (AWS)	7.91 / 2.60 = 3.0x	18
B200	192 GB	$3.69/hr	DeepInfra	$11.28/hr (Google Cloud)	11.28 / 3.69 = 3.1x	16
B300	288 GB	$4.89/hr	DeepInfra	$17.80/hr (AWS)	17.80 / 4.89 = 3.6x	12
GB200	192 GB	$8.00/hr	GMI Cloud	$16.00/hr (Oracle Cloud)	16.00 / 8.00 = 2.0x	3
A100 80GB	80 GB	$0.89/hr	DeepInfra	$3.22/hr (Google Cloud)	3.22 / 0.89 = 3.6x	18
L40S	48 GB	$0.47/hr	Vast.ai	$1.86/hr (AWS)	1.86 / 0.47 = 4.0x	13
MI300X	192 GB	$1.71/hr	TensorWave	none listed	n/a	4
RTX PRO 6000	96 GB	$0.93/hr	Vast.ai	$0.97/hr (Google Cloud)	0.97 / 0.93 = 1.0x	11
L4	24 GB	$0.11/hr	Vast.ai	$0.71/hr (Google Cloud)	0.71 / 0.11 = 6.5x	7
RTX 4090	24 GB	$0.14/hr	Vast.ai	none listed	n/a	4
RTX 5090	32 GB	$0.21/hr	Vast.ai	none listed	n/a	3

1. Hyperscalers charge 3.0x to 3.6x more for the same flagship silicon

For the data-center flagships, the cheapest hyperscaler rate is 3.0x the floor on H100 SXM ($5.38/hr vs $1.79/hr), 3.1x on B200 ($11.28/hr vs $3.69/hr) and 3.6x on B300 ($17.80/hr vs $4.89/hr). The widest gap in the table is L4 at 0.71 / 0.11 = 6.5x. You pay the hyperscaler premium for the ecosystem (IAM, networking, managed services, committed-use discounts, startup credits), which can be worth it, but it is a choice rather than a default. The exception is the new workstation-class Blackwell card: an RTX PRO 6000 starts at $0.93/hr (Vast.ai) and costs $0.97/hr on Google Cloud, almost the same.

2. Spot capacity is cheaper again, with two catches

If your job checkpoints, interruptible capacity cuts the price further: H100 SXM from $0.98/hr, H200 from $1.84/hr, B200 from $1.77/hr. The first catch: spot can be reclaimed at any moment, so it suits checkpointed training and offline batch jobs, never a user-facing endpoint. The second: the cheapest listings are often whole nodes. In the open dataset, the H200 spot floor on AWS requires 8 GPUs and the B200 spot floor on Google Cloud requires 8 GPUs, so the real entry ticket is that many times the per-GPU price.

3. Blackwell is rentable on demand

B200 is listed by 16 providers from $3.69/hr (DeepInfra), B300 by 12 from $4.89/hr (DeepInfra), and GB200 by 3 from $8.00/hr (GMI Cloud). If you assumed the newest silicon was contact-sales only, much of it now has a public hourly price.

4. For small models, consumer cards are hard to beat

An RTX 4090 (24 GB) rents from $0.14/hr and an RTX 5090 (32 GB) from $0.21/hr on marketplace hosts. 24 GB is enough for a 7B model in fp16 or a 13B model quantized to 4-bit (how much VRAM an LLM needs). Marketplace hosts are individual machines, so check host reliability and keep anything you cannot lose off the box.

5. The floor moves, so check it when you rent

Marketplace prices change as hosts come and go, and a cheap listing can be rented out within the hour. Each GPU page carries a 30-day price strip showing the low, the typical price and whether now is a good moment to rent (for example H200 or L40S).

How to choose, in four steps

Size the VRAM your model needs first. Everything else follows from it.
Decide whether the job tolerates interruption. Checkpointed training can use spot; an endpoint cannot.
Compare the landed cost, not the headline rate. Minimum GPU counts, egress, storage and cold-start billing can erase a cheap rate.
Check availability. A price with no stock behind it is not a price.

If you would rather skip the spreadsheet, describe the workload and FastGPU ranks the live options that can actually run it.

Get the data yourself

The prices behind this post are an open dataset (CC BY 4.0), downloadable as JSON or CSV with no key:

curl -s https://fastgpu.co/api/v1/dataset/gpu-prices-current.json \
  | jq '[.[] | select(.gpu_slug == "h100-sxm" and .offer_type == "on-demand")]
        | sort_by(.price_usd_hr) | .[:5]
        | map({provider: .provider_label, usd_per_gpu_hr: .price_usd_hr, min_gpus: .min_gpu_count})'

Daily history is in the same dataset, which also has a permanent DOI on Zenodo (10.5281/zenodo.22842387). If you are building an agent, the same data is exposed as an MCP server at https://fastgpu.co/api/mcp with list_gpu_prices and match_workload tools.

Method and disclosure

Prices are USD per GPU-hour, read from each provider's public API or pricing page and refreshed continuously. "Cheapest on-demand" counts on-demand and community listings, preferring ones that are not sold out; spot, reserved and serverless are shown separately. The hyperscaler reference is the cheapest on-demand offer for that card at AWS, Azure, Google Cloud or Oracle. Snapshot taken September 27, 2026 at 04:11 UTC; the linked pages show the current numbers. Some provider links on FastGPU are affiliate links; price tables and every "cheapest" figure here are ordered strictly by price.