--- title: "Run CUDA on a cloud GPU from $0.163/hr | PowerGPU" description: "NVIDIA CUDA on a fixed-price cloud GPU: one-click template, running in ~30 s, from $0.163/hr on RTX 4090. Best GPUs, volumes, per-second billing." url: https://powergpu.ai/templates/nvidia-cuda last_modified: 2026-09-14T11:04:20+00:00 prices_as_of: 2026-09-14 site: PowerGPU (powergpu.ai) --- Template · Base & frameworks · CUDA 13 # Run CUDA on a cloud GPU, in 30 seconds The bare CUDA base image — drivers, toolkit, SSH and Jupyter. Build your own stack on top. The bare CUDA image is the blank canvas: the NVIDIA driver stack, the CUDA 13 toolkit, nvcc, cuDNN, SSH and JupyterLab — nothing else. Use it to compile custom kernels, build your own framework stack, or reproduce a paper environment exactly, without fighting a pre-baked image. From **$0.163** /hr on an interruptible RTX 4090. NVIDIA CUDA powergpu/base:cuda13 (CUDA 13) (ARM) (SSH) (Jupyter) Pinned image, rebuilt on upstream releases. Override env, ports and disk at deploy. ### Running in ~30 s Pre-cached image, ports mapped, TLS terminated — the stack is working before you finish reading this. ### Template is free You pay the GPU price only: NVIDIA CUDA on a RTX 4090 is $0.327/hr on-demand, $0.163/hr interruptible, billed per second. ### Volumes for state Models, datasets and outputs live on a $0.08/GB/mo volume; the instance stays disposable. ### Private by default Dedicated GPU, encrypted disk, crypto payments, no KYC, no stored IPs — Jupyter and SSH behind your own credentials. ## Best GPUs for NVIDIA CUDA Three price points that run this template well — Good, Better, Best. Every model in the catalogue works; these are the value picks. | Tier | GPU | VRAM | On-demand | Interruptible | Why this card | | | --- | --- | --- | --- | --- | --- | --- | | (Good) | [RTX 4090](https://powergpu.ai/gpu/rtx-4090) | 24 GB | $0.327 | $0.163 | Cheap Ada card for compiling and testing kernels interactively. | [Deploy](https://cloud.powergpu.ai/?gpu=rtx-4090) | | (Better) | [A100 SXM4](https://powergpu.ai/gpu/a100-sxm4) | 80 GB | $0.560 | $0.280 | Datacenter Ampere with FP64 and MIG for library development. | [Deploy](https://cloud.powergpu.ai/?gpu=a100-sxm4) | | (Best) | [H100 SXM](https://powergpu.ai/gpu/h100-sxm) | 80 GB | $1.428 | $0.714 | Hopper features (FP8, TMA, thread-block clusters) for kernels that target the latest architecture. | [Deploy](https://cloud.powergpu.ai/?gpu=h100-sxm) | Need more VRAM? The [full catalogue](https://powergpu.ai/gpus) lists all 80 models with live availability; the [VRAM guide](https://powergpu.ai/guides/llm-vram-requirements) sizes models to cards. ## Deploy NVIDIA CUDA from the console, CLI or API Pick the template in the [console](https://cloud.powergpu.ai/) deploy bar, or script it: - **Console** — filter by GPU, choose *NVIDIA CUDA* in the template picker, set disk and env, deploy. - **CLI** — pip install powergpu, then the command on the right. [CLI reference](https://powergpu.ai/cli). - **API** — POST /v1/instances with "template": "nvidia-cuda". [REST reference](https://powergpu.ai/api). - **Own image** — any OCI reference works too; we inject the NVIDIA runtime. [Template docs](https://powergpu.ai/docs/templates). *deploy — nvidia-cuda* ``` $ powergpu launch --gpu rtx-4090 --template nvidia-cuda \ --disk 100 --volume models:/workspace/models ✓ instance i-7a41c0e2 running (27.9s) # NVIDIA CUDA · RTX 4090 · $0.327/hr · per second # https://i-7a41c0e2.powergpu.ai:8888 (TLS) $ powergpu stop i-7a41c0e2 # billing ends this second ``` ## What is inside the NVIDIA CUDA template - **Image**: powergpu/base:cuda13 - **CUDA**: CUDA 13 - **Access**: also builds for ARM hosts · SSH shell · JupyterLab on a mapped port - **Category**: [Base & frameworks](https://powergpu.ai/templates#base) - **Storage**: Instance NVMe disk (sized at deploy) + optional [network volumes](https://powergpu.ai/products/volumes) - **Billing**: GPU price only, per second — no template fee, no setup fee ## Other base & frameworks templates - [PyTorch on a cloud GPU](https://powergpu.ai/templates/pytorch) — The training and research default: PyTorch 2.6, cuDNN, JupyterLab and SSH out of the box. - [TensorFlow CUDA on a cloud GPU](https://powergpu.ai/templates/tensorflow-cuda) — TensorFlow with GPU support, Keras and TensorBoard on a mapped port. - [PyTorch NGC on a cloud GPU](https://powergpu.ai/templates/pytorch-ngc) — NVIDIA's NGC-optimised PyTorch build — tuned kernels for the datacenter cards. - [NVIDIA RAPIDS on a cloud GPU](https://powergpu.ai/templates/nvidia-rapids) — GPU-accelerated data science — cuDF, cuML, cuGraph in a notebook. - [All-in-One App Studio on a cloud GPU](https://powergpu.ai/templates/all-in-one-app-studio) — A launcher bundling the most-used AI apps behind one desktop — pick and run. - [All 37 templates](https://powergpu.ai/templates) ## NVIDIA CUDA on a cloud GPU: FAQ Environment variables, ports and custom images are covered in the [template docs](https://powergpu.ai/docs/templates). **How much does it cost to run NVIDIA CUDA on a cloud GPU?** Only the GPU price — the template is free. From $0.163/hr on an interruptible RTX 4090, $0.327/hr on-demand on a RTX 4090. Billing is per second, so an hour of tinkering costs an hour, not a day. Storage is $0.08/GB/month. **How long does NVIDIA CUDA take to start?** About 30 seconds from the deploy click: the image is pre-cached on hosts, ports are mapped and TLS is terminated for you. Restarting a stopped instance is faster, and your disk is exactly as you left it. **Can I keep my models and outputs between sessions?** Yes — attach a volume at deploy. Everything on the volume survives instance destruction and mounts on the next instance in the region in seconds, at $0.08/GB/month. The instance disk itself survives stop/start but not destroy. **Which CUDA version does the template ship?** CUDA 13 userspace on hosts running the current NVIDIA production driver. Templates pinned to CUDA 12.x exist for frameworks that need them; the offer card shows the maximum CUDA each machine supports. --- *About PowerGPU:* PowerGPU (powergpu.ai) is a cloud GPU rental service offering 80 NVIDIA GPU models — from the RTX A2000 at $0.024/hr to the B300 — at fixed prices set at least 30% below the public GPU marketplace median and re-checked weekly (H100 SXM: $1.428/hr on-demand). Billing is per second with no minimums; payment is crypto only (USDT, BTC, XMR, ETH, SOL, LTC, TRX) with no KYC. Instances run in Tier-III datacenters across 32 regions with a 99.9% uptime SLA and deploy in about 30 seconds from the web console (cloud.powergpu.ai) or the REST API. Source: https://powergpu.ai/templates/nvidia-cuda · Site index for AI assistants: https://powergpu.ai/llms.txt · Full content: https://powergpu.ai/llms-full.txt