AMD MI325X

Q: How much does the MI325X cost per hour?

As of September 5, 2026, the median on-demand price is $3.10 per GPU per hour across 3 providers. On-demand: 7 configs, median $3.10, cheapest $2.25 Reserved: 14 configs, median $1.95, cheapest $1.50 (12 mo) Spot: 1 config, median sold out Custom contract: 1 config, cheapest $2.25 No median for custom-contract: we only show this when at least 3 providers list the GPU on that billing type. Custom-contract pricing is negotiated per deal and usually not published.

Q: Who has the cheapest MI325X?

As of September 5, 2026, the lowest listed on-demand price for the MI325X is $2.25 per GPU per hour from Bentaus, though we haven't verified current stock.

Q: How much does the MI325X cost per month?

As of September 5, 2026, at 720 hours per month, one MI325X costs an estimated $2,232 at the median on-demand price.

Q: Where can I rent an MI325X?

GetDeploying currently tracks MI325X configs from 5 providers. See the full price comparison above for every provider and config.

Q: How many MI325X GPUs do I need?

One MI325X runs models up to roughly 408B parameters at 4-bit quantization or 110B at 16-bit, assuming a 32K context. Larger models run across multiple GPUs: the model table above shows the estimated memory and GPU count for popular open-weight LLMs.

Q: Why choose the MI325X?

256GB HBM3e with 6 TB/s bandwidth. Largest VRAM available for serving massive models without sharding. FP8 support for efficient inference.

Q: When is the MI325X not a good fit?

Very limited cloud availability. ROCm ecosystem, while rapidly improving, may have gaps that affect compatibility with some training frameworks. Consider Nvidia alternatives if CUDA compatibility is required.

Highest memory capacity AMD GPU for single-chip model serving.

Jump to: At a glance Pricing AI models Specs FAQ Alternatives

Compare vs other GPUs →

Aggregating historical prices...

Weekly median price per GPU per hour. AMD MI325X price history (JSON).

At a glance

Hardware

Architecture: CDNA 3
Memory per GPU: 256 GB HBM3e
Memory bandwidth: 6,000 GB/s
Release date: Q4 2024

Market Updated 7 minutes ago

Cheapest: $2.25 / GPU / hr; on-demand · not verified in stock
Median price: $3.10 / GPU / hr; on-demand
Coverage: 5 providers; 23 configs · none verified in stock

MI325X Pricing and Availability

5 providers

How this list works

The two views

By provider shows one card per company, placed where its best offer ranked. By configuration lists every offer.

Order

What you can rent comes first: in stock, then waitlist, then not reported, then out of stock. Priced offers rank ahead of quote-only, and on-demand ahead of other billing types.

Within a group, five factors set the order:

Location: datacenter proximity, blended with provider HQ.
Price: hourly price, per GPU and in total.
Billing type: reserved ahead of spot, spot ahead of quote-only.
Specs: more VRAM, vCPUs and RAM.
Provider diversity: a provider's repeat rows rank slightly lower, so one company can't take all the top spots.

Search

We match the provider, its country, GPU model, form factor, billing type, availability and instance name. Matching is partial and case-insensitive. Everyday words work too, so "interruptible" finds spot and "sold out" finds out of stock.

Transparency and funding

Ads and sponsors: Paid placements sit at the top of the list and are always labeled as sponsored content. Sponsorship never influences the organic ranking itself.
Affiliates: Affiliate links are marked. We may earn a commission if you click them, but commissions never affect the order.
Prices: Shown in USD, converted at daily reference rates where a provider publishes in another currency. A month means 720 hours. How we estimate costs.

DigitalOcean

USA 3 configs 1x-8x checked <24h ago

On-Demand from $3.80 Reserved from $2.88

From $3.80 / GPU / hr On-Demand Visit website


1x MI325X	256GB	--	--	Reserved (12mo)	$2.88	$2.88	Not reported
1x MI325X gpu-mi325x1-256gb	256GB	20	164GB	On-Demand	$3.80	$3.80	Not reported
8x MI325X gpu-mi325x8-2048gb	2.0TB	160	1.3TB	On-Demand	$3.80	$30.40	Not reported

In-stock and out-of-stock as of our last check at 2026-09-04T09:13Z. Not reported means we could not confirm availability.

Bentaus

USA 2 configs 8x checked 4 days ago

On-Demand from $2.25 Reserved from $1.95

From $2.25 / GPU / hr On-Demand Visit website


8x MI325X	2.0TB	--	--	Reservation	$1.95	$15.60	Not reported
8x MI325X	2.0TB	--	--	On-Demand	$2.25	$18.00	Not reported

In-stock and out-of-stock as of our last check at 2026-08-31T19:19Z. Not reported means we could not confirm availability.

Cyfuture AI

India 16 configs 1x-8x checked <24h ago

On-Demand from $3.06 Reserved from $1.50

From $3.06 / GPU / hr On-Demand Visit website

Configuration	Total VRAM	vCPUs	RAM	Billing	Per GPU/h	Total/h	Availability
8x MI325X 8xMI325.128v.2048m-12mo	1.5TB	128	1.5TB	Reserved (12mo)	$1.50	$11.98	Not reported
8x MI325X 8xMI325.128v.2048m	1.5TB	128	1.5TB	On-Demand	$3.06	$24.46	Not reported

In-stock and out-of-stock as of our last check at 2026-09-04T09:13Z. Not reported means we could not confirm availability.

Loading 14 more configs...

TensorWave

USA 1 config 8x checked 4 days ago

Custom from $2.25

From $2.25 / GPU / hr Custom OAM Visit website


8x MI325X OAM	2.0TB	128	3TB	Custom	$2.25	$18.00	Not reported

In-stock and out-of-stock as of our last check at 2026-08-31T22:15Z. Not reported means we could not confirm availability.

Vultr

Out of stock

USA 1 config 8x checked <15m ago

Spot from $2.00

From $2.00 / GPU / hr Spot Visit website


8x MI325X vbm-256c-3072gb-8-mi325x-gpu	2.0TB	256	3.1TB	Spot	$2.00	$16.00	Out of stock

In-stock and out-of-stock as of our last check at 2026-09-05T07:02Z. Not reported means we could not confirm availability.

Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Verify before provisioning. More on how we price.

What size AI models can the MI325X run?

One MI325X has 256 GB of VRAM. In practice, that's enough memory for roughly 408B parameters at 4-bit or 110B at 16-bit, assuming a 32K context. Below are some open-weight LLMs, with the estimated memory and GPUs each one needs.

Model	Parameters	4-bitINT4 / FP4	8-bitFP8 / INT8	16-bitFP16 / BF16
Qwen3.8-27B	27B	1 GPU 18 GB VRAM · $2,232/mo	1 GPU 30 GB VRAM · $2,232/mo	1 GPU 58 GB VRAM · $2,232/mo
Gemma 4 31B	30.7B	1 GPU 22 GB VRAM · $2,232/mo	1 GPU 35 GB VRAM · $2,232/mo	1 GPU 69 GB VRAM · $2,232/mo
GPT-OSS-120B	117B 5.1B active	1 GPU 67 GB VRAM · $2,232/mo	1 GPU 120 GB VRAM · $2,232/mo	8 GPUs 237 GB VRAM · $17,856/mo
DeepSeek V4 Flash	284B 13B active	1 GPU 160 GB VRAM · $2,232/mo	8 GPUs 287 GB VRAM · $17,856/mo	8 GPUs 573 GB VRAM · $17,856/mo
GLM-5.3-Flash	320B 18B active	1 GPU 178 GB VRAM · $2,232/mo	8 GPUs 322 GB VRAM · $17,856/mo	8 GPUs 642 GB VRAM · $17,856/mo
MiniMax-M3	428B 23B active	8 GPUs 239 GB VRAM · $17,856/mo	8 GPUs 432 GB VRAM · $17,856/mo	8 GPUs 862 GB VRAM · $17,856/mo
Kimi K3	2.8T 104B active	8 GPUs 1,544 GB VRAM · $17,856/mo	13 GPUs 2,804 GB VRAM	25 GPUs 5,605 GB VRAM

Estimates based on the median on-demand rate. Memory is weights plus FP8 KV cache at 32K context per request (FP16/BF16 in the 16-bit column). GPU counts assume 90% of advertised VRAM is usable. No guarantee of runtime support, usable performance, or that a matching quantized build exists. How we estimate costs.

Technical Specifications

AMD MI325X · Per GPU

Compute · dense
FP8	2,614.9 TFLOPS 5,229.8 with sparsity
FP16 / BF16	1,307.4 TFLOPS 2,614.9 with sparsity
INT8	2,614.9 TOPS 5,229.8 with sparsity
FP32	163.4 TFLOPS
FP64	81.7 TFLOPS
Precision support	FP8FP16BF16TF32FP32FP64INT8

Memory
Capacity	256 GB HBM3e
Bandwidth	6,000 GB/s
Bus width	8,192-bit
ECC	Yes

Silicon
Architecture	CDNA 3
Process	TSMC 5nm / 6nm FinFET
Transistors	153 billion
Shader cores	19,456 Stream processors
Matrix cores	1,216
Compute units	304

Fabric and host
GPU interconnect	Infinity Fabric 896 GB/s
Host interface	PCIe 5.0 x16

Power
Board power	1,000 W
Cooling	Passive

Platform
Partitioning	Up to 8 partitions

Source: official AMD MI325X datasheet.

Runpod

Our sponsor

Spin Up a GPU in Seconds - No Setup Required

Deploy AI Models Globally in 30+ Regions

Scale from 0 to 1000 GPUs on Demand

PyTorch, Tensorflow & Docker Ready

Learn more

Frequently Asked Questions

How much does the MI325X cost per hour?

As of September 5, 2026, the median on-demand price is $3.10 per GPU per hour across 3 providers.

Billing type	Configs	Median / GPU / hr	Cheapest in stock
On-demand	7	$3.10	$2.25
Reserved	14	$1.95	$1.50 (12 mo)
Spot	1	Sold out
Custom contract	1		$2.25

No median for custom-contract: we only show this when at least 3 providers list the GPU on that billing type. Custom-contract pricing is negotiated per deal and usually not published.

Who has the cheapest MI325X?

As of September 5, 2026, the lowest listed on-demand price for the MI325X is $2.25 per GPU per hour from Bentaus, though we haven't verified current stock.

How much does the MI325X cost per month?

As of September 5, 2026, at 720 hours per month, one MI325X costs an estimated $2,232 at the median on-demand price.

Where can I rent an MI325X?

GetDeploying currently tracks MI325X configs from 5 providers. See the full price comparison above for every provider and config.

How many MI325X GPUs do I need?

One MI325X runs models up to roughly 408B parameters at 4-bit quantization or 110B at 16-bit, assuming a 32K context. Larger models run across multiple GPUs: the model table above shows the estimated memory and GPU count for popular open-weight LLMs.

Why choose the MI325X?

256GB HBM3e with 6 TB/s bandwidth. Largest VRAM available for serving massive models without sharding. FP8 support for efficient inference.

When is the MI325X not a good fit?

Very limited cloud availability. ROCm ecosystem, while rapidly improving, may have gaps that affect compatibility with some training frameworks. Consider Nvidia alternatives if CUDA compatibility is required.

Alternatives to AMD MI325X

Browse all GPUs

Last updated Sept. 5, 2026

AMD MI325X

At a glance

Hardware

Market Updated 7 minutes ago

MI325X Pricing and Availability

How this list works

The two views

Order

Search

Transparency and funding

DigitalOcean

Bentaus

Cyfuture AI

TensorWave

Vultr

What size AI models can the MI325X run?

Technical Specifications

Frequently Asked Questions

How much does the MI325X cost per hour?

Who has the cheapest MI325X?

How much does the MI325X cost per month?

Where can I rent an MI325X?

How many MI325X GPUs do I need?

Why choose the MI325X?

When is the MI325X not a good fit?

Alternatives to AMD MI325X

Nvidia H100

Nvidia B200

Nvidia H200

AMD MI300X