GPU Cloud

RTX 5090, dedicated.

A single-GPU RTX 5090 machine, 32 GB of GDDR7 on NVIDIA's Blackwell architecture, on an x86 host. Rentable now at $0.89 per hour, billed per minute, the same terms as every other GPUwerk tier. It runs in GPUwerk's EU-Central data center in the Czech Republic, and you connect over SSH with your own key. Terminate when done and billing ends.

Rentable now, one machine. GPUwerk has a single RTX 5090, rented first come, first served. Deploy it when it is free; while it is in use, join its waitlist and GPUwerk emails you when it frees up.

Specifications

Per node. Every GPUwerk instance, on any tier, is one whole machine to yourself.

NVIDIA GeForce RTX 5090
GPURTX 5090, Blackwell architecture
Memory32 GB GDDR7
Hostx86-64 Intel host; up to 48 GB of system RAM for your instance
TenancyOne customer per machine, never shared

No tok/s or training throughput figures published yet: GPUwerk publishes figures only after running them on its own hardware, and the RTX 5090 runs are not published yet. See the DGX Spark fleet benchmarks for how they are reported. Third-party figures for this card are collected in RTX 5090 LLM benchmarks; which models fit in 32 GB is in best LLMs for an RTX 5090, and training limits in fine-tuning on an RTX 5090.

What would you run on it?

32 GB of fast VRAM is the consumer sweet spot for a single model that fits without heavy quantization.

Image and video generation

Diffusion pipelines that want raw VRAM bandwidth over unified memory capacity: the case where a discrete GPU usually beats a Spark.

Mid-size model serving

A 13B to 32B-class model at Q4 to Q8 fits comfortably, with room for a reasonable batch size.

Fast iteration

Consumer-tier CUDA compute for a single developer's fine-tuning or eval loop, where the 128 GB of a Spark is more capacity than the job needs.

Billing

Rate$0.89 per hour
BillingPer minute, against a prepaid credit balance
FeesNo egress fees
CommitmentNone

Prices are in USD, excluding applicable tax. Full billing mechanics, including what a stop versus a terminate does, are on the pricing page.

FAQ

Can I rent an RTX 5090 now?

Yes. GPUwerk has one RTX 5090 machine, rented first come, first served at $0.89 per hour. Sign in, pick the RTX 5090 at deploy and add an SSH key; you get your own public IP with SSH. When it is in use, join its waitlist and GPUwerk emails people in line one at a time when it frees up.

What's the host configuration around the GPU?

The host configuration (CPU, system RAM, storage) of the installed machine is shown in the console when you deploy. GPUwerk publishes figures on this page only after measuring them on its own hardware.

How does billing work on this tier?

The same as every GPUwerk tier: per minute, against a prepaid credit balance, in USD, excluding applicable tax, with no egress fees and no commitment.

Need GPU time today?

A DGX Spark has 128 GB of unified memory and is rentable now, at $0.79/hour billed per minute. First top-up: pay $20, get $30 in credit. See how a Spark compares with a discrete card, or deploy an RTX 5090.