🚀 One API for every frontier model on FastInfra

Browse models

Transparent per-token pricing → compare providers on FastInfra

View pricing

OpenAI-compatible chat completions → start in minutes

Read the docs

⚡ Server Auction → Enterprise bare metal from $78.15/mo (173 in stock)

Browse deals

AMD Instinct MI350X cloud GPU rental

Rent an AMD Instinct MI350X cloud GPU from $7.41 per hour with 294 GB of VRAM. FastInfra bills pods per minute with no commitment, so a one-hour experiment costs $7.41 and you stop paying the moment you delete the pod. Run 1 to 8 GPUs per pod, connect over SSH or JupyterLab, and see the exact price before you deploy.

Secure $7.41 / GPU-hr 294 GB VRAM

Prices synced November 22, 2026 · billed per minute

At a glance

MI350X specs and pricing

On-demand from
$7.41
per GPU-hour
GPU memory
294 GB
VRAM
GPUs per pod
1–8
2352 GB max VRAM

The AMD Instinct MI350X is a GPU with 294 GB of VRAM. On FastInfra it is a good fit for frontier-scale training, 200B+ inference, long-context serving. As a rule of thumb, one MI350X holds the weights of a model of about 110B parameters at 16-bit precision, or about 470B with 4-bit quantization, after leaving room for the KV cache.

The MI350X runs on the Secure tier in vetted datacenters at $7.41 per GPU-hour.

Pricing

MI350X price per hour by GPU count

Live FastInfra prices per pod-hour. Pods bill per minute, so you pay only for the minutes a pod exists.

MI350X hourly price by number of GPUs
Pod sizeTotal VRAM Secure / hr Month (730 h, lowest)
1× MI350X 294 GB $7.41 $5,411
2× MI350X 588 GB $14.82 $10,822
4× MI350X 1176 GB $29.65 $21,643
8× MI350X 2352 GB $59.30 $43,286

What an MI350X costs over time

One GPU at the lowest on-demand rate ($7.41/hr).

MI350X cost by duration
DurationOn-demand
1 hour $7.41
8-hour working day $59.30
24 hours $178
1 week (168 hours) $1,245
1 month (730 hours) $5,411
Model fit

Which LLMs fit on the MI350X?

GPUs needed to hold the model weights with about 20% of VRAM left for the KV cache. Rule of thumb only: long contexts and big batches need more.

MI350X GPUs needed per model size and precision
Model sizeExamplesFP16 / BF168-bit4-bit
7–8BLlama 3.1 8B, Qwen3 8B, Mistral 7B1 GPU1 GPU1 GPU
12–14BGemma 3 12B, Qwen3 14B, Phi-41 GPU1 GPU1 GPU
27–32BGemma 3 27B, Qwen3 32B1 GPU1 GPU1 GPU
70BLlama 3.3 70B1 GPU1 GPU1 GPU
120BGPT-OSS 120B2 GPUs1 GPU1 GPU
235BQwen3 235B-A22B2 GPUs1 GPU1 GPU
405BLlama 3.1 405B4 GPUs2 GPUs1 GPU
671BDeepSeek V3 / R18 GPUs4 GPUs2 GPUs

Best for

Frontier-scale training, 200B+ inference, long-context serving.

Only need inference?

If you want to call an open model such as Llama, Qwen, DeepSeek or GPT-OSS rather than manage a server, the pay-per-token inference API is often cheaper than keeping a GPU busy around the clock. It is OpenAI-compatible and uses the same account and wallet as your pods. Browse every model with live prices.

Alternatives

MI350X vs similar GPUs

Live lowest on-demand price per GPU-hour, and price per GB of VRAM.

GPUs similar to the MI350X
GPUVRAMArchitectureFrom / hrPer GB-hr
MI350X (this page) 294 GB — $7.41 $0.025
NVIDIA B300 288 GB Blackwell $9.37 $0.033
AMD Instinct MI300X 192 GB CDNA 3 $3.23 $0.017
NVIDIA B200 180 GB Blackwell $8.07 $0.045
NVIDIA H200 NVL 143 GB Hopper $5.12 $0.036
NVIDIA H200 SXM 141 GB Hopper $4.85 $0.034
NVIDIA RTX PRO 6000 Max-Q 96 GB Blackwell $2.21 $0.023
FAQ

MI350X rental — frequently asked questions

How much does it cost to rent an AMD Instinct MI350X per hour?

$7.41 per GPU-hour on demand. Billing is per minute while the pod exists, so 24/7 use for a 730-hour month costs about $5,411 per GPU at the lowest on-demand rate. Prices sync from live capacity and are shown before you deploy.

How much VRAM does the AMD Instinct MI350X have?

294 GB of VRAM. A pod with 8 GPUs gives 2,352 GB of combined VRAM for tensor- or pipeline-parallel workloads.

What size LLM can I run on an AMD Instinct MI350X?

Roughly 110B parameters at FP16 or BF16, 230B at 8-bit and 470B at 4-bit on a single GPU, reserving about 20% of VRAM for the KV cache and activations. Mixture-of-experts models need memory for all experts, not only the active ones. Long contexts and large batches need more headroom.

Is the AMD Instinct MI350X billed per hour or per minute?

Per minute. The hourly rate is divided by 60 and charged for every minute the pod exists. Attached volumes bill per GB-month and continue while a pod is stopped; delete the pod to stop all charges. There is no minimum term or reservation fee.

How do I rent an AMD Instinct MI350X on FastInfra?

Create an account, add prepaid credit, open the GPU configurator and choose the MI350X, the tier, the GPU count, a container template and storage. The configurator shows the hourly and monthly price, then deploys a pod you reach over SSH or JupyterLab within minutes.

Can I rent multiple MI350X GPUs in one pod?

Yes, up to 8 per pod. The price scales linearly: 8 GPUs cost $59.30 per hour on Secure.

Deploy an MI350X in minutes

Price the exact pod, storage and template in the configurator, add credit, and connect over SSH or JupyterLab. Billed per minute, no commitment.