On-Demand VMs

A machine, not a container.

Full root over the whole machine: your own Docker, your own kernel, your own Kubernetes. 1 to 8 NVIDIA GPUs, live in about 90 seconds, billed by the minute.

NVIDIA H200 from $3.99/hrH100 from $2.69/hrper minute, no minimum rental

Yours, down to the kernel.

Take root with sudo. Patch the kernel, load your own modules, mount what you like.

Your orchestrator, not ours.

Your own Docker daemon, or the box joined to your cluster as a GPU node.

1 to 8 GPUs on one box.

Same image, same command, more cards.

What is an on-demand GPU VM?

Most GPU clouds rent you a container. You get a process namespace on a host somebody else configured, with the kernel, the driver version and the container runtime already decided for you. That is fine right up to the moment your work needs something the host does not offer: a different driver, a custom kernel module, a privileged mount, your own container runtime, a kubelet.

An on-demand VM is the machine itself. You get root over one to eight dedicated NVIDIA GPUs, a full operating system you are free to replace, and a network you can reach on your own terms. Docker is something you install, not something you are handed. Kubernetes is something you join, not something you ask us to enable.

Billing does not change: per minute, no minimum rental period, and pausing stops the GPU charge.

Configurations

One card, or eight of them.

Pick the GPU, then the count. Rates below are for the whole instance, on demand, billed by the minute.

GPUTotal VRAMvCPURAMOn demand
NVIDIA H200 SXM141 GB28300 GB$3.99/hr
NVIDIA H100 SXM80 GB16200 GB$2.69/hr
NVIDIA RTX Pro 6000 Blackwell96 GB28160 GB$1.89/hr
NVIDIA A100 80GB80 GB16112 GB$1.49/hr
NVIDIA A100 40GB40 GB16112 GB$0.89/hr
NVIDIA A3024 GB16112 GB$0.41/hr
NVIDIA L424 GB32124 GB$0.44/hr

vCPU and system RAM scale with the number of GPUs in the instance.

Instance storage is $0.00014/GB/hour. A paused instance is charged storage only.

Runtime

Bring your own runtime.

The VM ships with GPU drivers and nothing you did not ask for. What runs on top is your call.

Docker

Install the daemon at the version your stack was built against. Pull from your own registry, including private ones. Use the NVIDIA container toolkit, or go around it.

Kubernetes

Join the instance to an existing cluster as a GPU node, or stand up a single-node cluster on the box. We do not run your control plane, which means we cannot break it.

Anything else

containerd, Podman, Apptainer, LXC, or no container at all. It is your kernel, so the list is as long as the kernel allows.

Choosing

VMs or containers?

Same GPUs underneath, at the same hourly rate. The difference is how much of the stack you want to own, and how fast you want to be working.

On-Demand VMThis pageTemplate (managed container)
AccessRoot on the machineRoot inside the container
Kernel and driversYours to choose and patchFixed by the host
DockerInstall your own daemonNot available inside
KubernetesJoin or host a clusterNot available
Time to first shellabout 90 seconds1.8 seconds
Pre-built stacksYou build the imagePyTorch, ComfyUI and more, ready
How you reach itSSHJupyterLab, VS Code in the browser, or SSH
GPUs per instance1 to 81 to 8
Storage across stop and startPersistsPersists
BillingPer minutePer minute
Pick this whena container is not a machineyou want to start training right away
Benchmarks

Does the VM slow the GPU down?

Not in our tests. Internal 8× H200 runs put VM and container results within 1.2% of each other across tensor compute, HBM bandwidth, NCCL communication and training throughput. Five of the six were within 0.5%.

A virtual machine adds a hypervisor and a guest OS, and buyers reasonably assume the GPUs pay for it. They do not have to: the physical H200s are assigned to the VM through direct passthrough, CUDA executes on the card itself, and NVLink and NVSwitch still connect the eight GPUs. That is the architecture. Below is the measurement.

Training throughput, nanochat, 8 GPUs
Container0tok/s
VM0tok/s
+0.26% observed differenceFinal 10-step mean. FP8, Flash Attention 3.
NCCL all-reduce bus bandwidth
Container0.0GB/s
VM0.0GB/s
-0.17% observed difference1 GiB message across 8 GPUs. The NVLink and NVSwitch path, preserved.
What we measuredContainerVMVM difference
BF16 matrix multiply, mean of 8 GPUs694.1 TFLOP/s695.9 TFLOP/s+0.25%
FP16 matrix multiply, mean of 8 GPUs623.0 TFLOP/s630.3 TFLOP/s+1.16%
HBM copy bandwidth, mean of 8 GPUs4.286 TB/s4.266 TB/s-0.47%
NCCL all-reduce bus bandwidth, 1 GiB message468.2 GB/s467.4 GB/s-0.17%
nanochat training, 1 GPU205,133 tok/s206,131 tok/s+0.49%
nanochat training, 8 GPUs1,594,318 tok/s1,598,385 tok/s+0.26%

Differences in either direction at this scale are run-to-run variation: clocks, temperature and software builds move results by fractions of a percent. Read the table as parity, not as the VM winning four rows of six.

Internal JarvisLabs benchmark, June 2026, run on separate 8× NVIDIA H200 SXM allocations. A JarvisLabs GPU container compared with a KVM VM using direct GPU passthrough. Training was nanochat with FP8 and Flash Attention 3; probe software versions differed slightly between runs. These figures describe the tested systems and workloads, not a universal guarantee.

How it works

Four steps, then it is your box.

01

Pick a GPU, a count and a region

From the dashboard, or with one jl create.

02

SSH in and take root

About 90 seconds later. Drivers are loaded, sudo is yours, and the rest of the machine is empty.

03

Pause when you stop

GPU charges stop with it. You pay $0.00014/GB/hour for storage while it sits.

04

Resume onto the same disk

Your data, your images and your environment are where you left them.

Platform

What every VM sits on.

The same platform under every instance, whichever GPU you picked and however many of them.

your workload
Compute & network
Dedicated GPUsPrivate networkingNetwork file storage

Your GPUs are never shared. On VMs you control the kernel, drivers and OS. Your own isolated VPC with IP ranges you choose. One filesystem, attached to as many of your own instances as you need.

Team & billing
Teams and billing

Admins manage members, allocate credits, and see team usage.

Where it runs
Data residency

Choose where your data lives. India and Europe today, more regions coming soon.

In use

What people run on them.

Workloads researchers actually run, and the card that handles each one.

WorkloadRuns well onFrom
Training from scratchNVIDIA H100 SXM$2.69/hr
RL post-training and GRPONVIDIA A100-80GB$1.49/hr
Long-running experiment sweepsNVIDIA L4$0.44/hr
Fine-tuning open modelsNVIDIA A100-80GB$1.49/hr
CLI and SDK

Every VM, from the terminal.

Create, connect, copy files and tear down without opening the dashboard. Your agents can drive it too.

Read the CLI docs
Questions

Answers.

Can't find what you're looking for? Reach out to our support team.

Yes. SSH in and take root with sudo: install what you like, patch the kernel, load your own modules. It is a virtual machine, not a namespace on somebody else’s host.

Yes, your own daemon at your own version, pulling from your own registry. The NVIDIA container toolkit is yours to install or skip.

Yes. The VM behaves as a GPU node: install a kubelet and join an existing cluster, or run a single-node cluster on the box. We do not operate your control plane.

Not measurably, for the GPU paths we tested. In internal 8× H200 runs, VM and container results stayed within 1.2% of each other across tensor compute, HBM bandwidth, NCCL communication and nanochat training throughput, and five of the six results were within 0.5%. The GPUs reach the VM by direct passthrough, so CUDA runs on the card itself.

About 90 seconds to a root shell. If you want a working PyTorch environment in 1.8 seconds instead, that is what Templates are for.

GPU charges stop the moment you pause. You pay storage only, at $0.00014/GB/hour, and your disk is kept.

No. Your GPUs are dedicated to your instance for as long as it runs, and on a VM you control the kernel, the drivers and the OS on top of them.

Eight is the maximum in a single instance. For anything larger, talk to us.

Not every GPU is offered at every count. NVIDIA H200 SXM, NVIDIA H100 SXM, NVIDIA RTX Pro 6000 Blackwell, NVIDIA L4 run at 8 per instance. The rest are capped lower, and the configurations table shows exactly which.

A $10 minimum to add credit. Billing is per minute with no minimum rental period. There is no free trial.

Get started

Start building on The AI Cloud.