NVIDIA RTX 5090 cloud GPU rental

The NVIDIA RTX 5090 is a consumer Blackwell GPU with 32 GB of GDDR7 at 1.8 TB/s of memory bandwidth. On FastInfra it is a good fit for 7B–13B inference, LoRA fine-tuning, ComfyUI and Stable Diffusion. As a rule of thumb, one RTX 5090 holds the weights of a model of about 12B parameters at 16-bit precision, or about 51B with 4-bit quantization, after leaving room for the KV cache.

Two on-demand tiers are available. Secure runs in vetted datacenters at $1.29 per GPU-hour; Value uses lower-cost community hosts at $0.746 per GPU-hour for the same GPU model, with more variable capacity.

Best for

7B–13B inference, LoRA fine-tuning, ComfyUI and Stable Diffusion.

About the Blackwell architecture

RTX Blackwell cards pair fifth-generation Tensor Cores (with FP4 and FP8 support) with GDDR7 memory. They give much of the inference throughput of datacenter parts at a lower hourly price, and suit LLM serving with quantized weights, image and video generation, and fine-tuning of small and mid-size models. Use a CUDA 12.8+ image for full Blackwell support.

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.

How much does it cost to rent an NVIDIA RTX 5090 per hour?

$1.29 per GPU-hour on the Secure tier and $0.746 on the Value tier. Billing is per minute while the pod exists, so 24/7 use for a 730-hour month costs about $545 per GPU at the lowest on-demand rate. Prices sync from live capacity and are shown before you deploy.

How much VRAM does the NVIDIA RTX 5090 have?

32 GB of GDDR7 at 1.8 TB/s of memory bandwidth. The chip is rated at up to 575 W. A pod with 8 GPUs gives 256 GB of combined VRAM for tensor- or pipeline-parallel workloads.

What size LLM can I run on an NVIDIA RTX 5090?

Roughly 12B parameters at FP16 or BF16, 25B at 8-bit and 51B 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 NVIDIA RTX 5090 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 NVIDIA RTX 5090 on FastInfra?

Create an account, add prepaid credit, open the GPU configurator and choose the RTX 5090, 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 RTX 5090 GPUs in one pod?

Yes, up to 8 per pod. The price scales linearly: 8 GPUs cost $10.29 per hour on Secure.

Is there a cheaper GPU than the RTX 5090 with the same VRAM?

Yes. The NVIDIA RTX 5000 Ada has 32 GB and starts at $0.53 per hour. It differs in architecture (Ada Lovelace), memory bandwidth and multi-GPU interconnect, so compare throughput for your workload before switching.