AMD MI355X

AMD MI355X

AMD's next-gen CDNA 4 competitor to Nvidia Blackwell.

Arch
CDNA 4
Memory
288 GB HBM3e
Bandwidth
8.00 TB/s
Released
Q2 2025

Price history

Weekly median price per GPU per hour · Get the data

Last 12 months
Aggregating historical prices...

At a glance Updated 9 minutes ago

Cheapest on-demand
$5.75 /GPU/hr
not verified in stock · 8x · Bentaus
Median price (current)
$7.28 /GPU/hr
on-demand
90-day trend
--
Coverage
10 providers
21 price points

Pricing & availability

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:

  1. Location: datacenter proximity, blended with provider HQ.
  2. Price: hourly price, per GPU and in total.
  3. Billing type: reserved ahead of spot, spot ahead of quote-only.
  4. Specs: more VRAM, vCPUs and RAM.
  5. 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. Pricing methodology.
Daytona logo

Daytona

In stock

10 configs 1x-8x

On-Demand
from $6.20
Spot
from $3.56
From $6.20 /GPU/hr On-Demand
OpenRelay logo

1 config 1x

On-Demand
from $6.30
From $6.30 /GPU/hr On-Demand
Bentaus logo

2 configs 8x

On-Demand
from $5.75
Reserved
from $5.25
From $5.75 /GPU/hr On-Demand
Oracle Cloud logo

1 config 8x

On-Demand
from $8.60
From $8.60 /GPU/hr On-Demand
TensorWave logo

1 config 8x

Custom
from $2.95
From $2.95 /GPU/hr Custom OAM
Cirrascale logo

1 config 8x

Reserved
from $4.29
From $4.29 /GPU/hr Reserved
DigitalOcean logo

1 config 8x

Spot
from $4.50
From $4.50 /GPU/hr Spot
Impossible Cloud logo

2 configs 8x

Reserved
from $6.13
From $6.13 /GPU/hr Reserved (24mo)
Crusoe logo

1 config 8x

Custom
on request
Pricing On request Custom
Vultr logo

Vultr

Out of stock

1 config 8x

Spot
from $2.59
From $2.59 /GPU/hr Spot
Visit website Affiliate link

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 can the MI355X run?

One MI355X has 288 GB of VRAM. In practice, that's enough memory for roughly 460B parameters at 4-bit or 125B at 16-bit, assuming a 32K context. Below are some open-weight LLMs, with the estimated memory and GPUs each one needs.

Fits on 1 Single node ≤ 8 Multi-node
Model Memory (INT4 / FP4) MI355Xs needed Cost /hr Cost /mo
Alibaba Cloud logo Qwen3.8-27B 27B
18 GB
1
$7.28
$5,242
Google Cloud logo Gemma 4 31B 30.7B
22 GB
1
$7.28
$5,242
OpenAI logo GPT-OSS-120B 117B · 5.1B active
67 GB
1
$7.28
$5,242
Z.AI logo GLM-5.3-Flash 320B · 18B active
178 GB
1
$7.28
$5,242
MiniMax logo MiniMax-M3 428B · 23B active
239 GB
1
$7.28
$5,242
DeepSeek logo DeepSeek V4.1 Flash 552B · 16B active
306 GB
8
$58.24
$41,933
Moonshot AI logo Kimi K3 2.8T · 104B active
1,544 GB
8
$58.24
$41,933

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. Pricing methodology.

Technical Specifications

AMD MI355X · Per GPU

Compute · dense
FP4 10,066.3 TFLOPS
FP8 5,033.2 TFLOPS 10,066.4 with sparsity
FP16 / BF16 2,516.6 TFLOPS 5,033.2 with sparsity
INT8 5,033.2 TOPS 10,066.4 with sparsity
FP32 157.3 TFLOPS
FP64 78.6 TFLOPS
Precision support FP4FP6FP8FP16BF16FP32FP64INT8
Memory
Capacity 288 GB HBM3e
Bandwidth 8,000 GB/s
Bus width 8,192-bit
ECC Yes
Silicon
Architecture CDNA 4
Process TSMC 3nm / 6nm FinFET
Transistors 185 billion
Shader cores 16,384 Stream processors
Matrix cores 1,024
Compute units 256
Fabric and host
GPU interconnect Infinity Fabric 1,075 GB/s
Host interface PCIe 5.0 x16
Power
Board power 1,400 W
Platform
Partitioning Up to 8 partitions

Source: official AMD MI355X datasheet.

Frequently Asked Questions

How much does the MI355X cost per hour?

As of November 19, 2026, the median on-demand price is $7.28 per GPU per hour across 4 providers.

Billing typeConfigsMedian /GPU/hrCheapest

On-demand

8

$7.28

$5.75 (not verified in stock, Bentaus)

Reserved

4

$5.25

$4.29 (not verified in stock, Cirrascale)

Spot

7

$3.56 (in stock, Daytona)

No median for spot: we only show this when at least 3 providers list the GPU on that billing type. 2 providers also quote the MI355X on a custom contract, priced per deal.

Who has the cheapest MI355X?

The lowest listed on-demand price for the MI355X is $5.75 per GPU per hour from Bentaus, though we haven't verified current stock. Cheapest verified in stock: on-demand $6.20 from Daytona. Other billing types: reserved $4.29 (not verified in stock) from Cirrascale and spot $3.56 (in stock) from Daytona.

How much does the MI355X cost per month?

At 720 hours per month, one MI355X costs an estimated $5,242 at the median on-demand price. Cheapest verified in stock: $4,464 per month on-demand, $2,563 spot.

Where can I rent an MI355X?

GetDeploying currently tracks MI355X configs from 10 providers. Daytona has verified in-stock on-demand configs. See the full price comparison above for every provider and config.

How many MI355X GPUs do I need?

One MI355X runs models up to roughly 460B parameters at 4-bit quantization or 125B 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 MI355X?

288GB HBM3e with FP4 support on CDNA 4 architecture. Next-generation AMD compute for both training and inference. Competitive with Nvidia Blackwell on memory capacity.

When is the MI355X not a good fit?

Latest generation with minimal cloud availability. ROCm software stack, while rapidly improving, may lag behind CUDA for some workloads. Evaluate software compatibility before committing.

Alternatives to AMD MI355X

Last updated