The marketplace for on-demand GPUs

The GPU cloud
to train.

Stop configuring datacenters and start computing. The cheapest GPUs from every cloud, behind one clean workflow. Launch a machine in seconds, pay by the second.

441GPUs live nowACROSS 2 CLOUDS
$0.14starting per hourBILLED BY THE SECOND
2clouds aggregatedONE UNIFIED FEED
~90sto a live boxKEY INJECTED AT BOOT
441GPUs live nowACROSS 2 CLOUDS
$0.14starting per hourBILLED BY THE SECOND
2clouds aggregatedONE UNIFIED FEED
~90sto a live boxKEY INJECTED AT BOOT
Capabilities

Everything you need.
Nothing you don't.

01

Live in ninety seconds

Pick a GPU and we provision it on the cheapest acceptable cloud, then hand you a ready-to-run SSH command. No tickets, no quotas, no waiting room.

02

Every cloud, one feed

We aggregate multiple GPU clouds into a single price-sorted feed. You never pick a provider; we route to the best box and hand you the connection.

03

Private by tunnel

The rented box reverse-tunnels its SSH to our gateway. The provider's real IP is never exposed, to you or to anyone scanning the network.

boxipx
04

Your key, your box

We inject your SSH key at boot, so only you can log in. Dedicated hardware, never shared tenancy, torn down clean the moment you are done.

Process

Three steps.
Then nvidia-smi.

ipx.session
1# one feed, every cloud, cheapest first
2GPU         VRAM    $/hr    region
3H100 SXM    80 GB   1.89    us-east
4RTX 4090    24 GB   0.34    eu-west
5# click rent on the one you want
1# the box dials home, not the other way
2rented box  -->  gpu.ipxtechholding.com:20137
3# the provider ip is never exposed
4status: tunnel established
5status: your key authorized
1# your key, your box, one hostname
2$ ssh -p 20137 root@gpu.ipxtechholding.com
3root@gpu:~# nvidia-smi
4# tear down anytime, billed by the second
5root@gpu:~# exit
Ready
Network

Every cloud,
one network.

We watch every connected cloud and refresh the whole feed every minute. You see one price-sorted list and rent in a click, while we handle the provider behind the gateway.

441
GPUs live
2
Clouds aggregated
$0.14
From, per hour
Live availabilityAll operational
80GB+ VRAM
91 offers live
from $2.54/hr
40 to 79GB VRAM
59 offers live
from $0.49/hr
24 to 39GB VRAM
178 offers live
from $0.22/hr
Under 24GB VRAM
113 offers live
from $0.14/hr
Live metrics

Performance you
can measure.

Live|--:--:--
0
GPUs live now
0
Clouds aggregated
0
Feed refresh interval
0
Seconds to a live box
Hardware

Works with the GPUs
you actually want.

From 24GB consumer cards to 80GB datacenter accelerators, across two clouds.

H100 SXM
80GB HBM3
H100 PCIe
80GB
A100
80GB
A100
40GB
L40S
48GB
RTX 4090
24GB
RTX 6000 Ada
48GB
RTX 3090
24GB
A6000
48GB
V100
32GB
L4
24GB
A40
48GB
H100 SXM
80GB HBM3
H100 PCIe
80GB
A100
80GB
A100
40GB
L40S
48GB
RTX 4090
24GB
RTX 6000 Ada
48GB
RTX 3090
24GB
A6000
48GB
V100
32GB
L4
24GB
A40
48GB
A40
48GB
L4
24GB
V100
32GB
A6000
48GB
RTX 3090
24GB
RTX 6000 Ada
48GB
RTX 4090
24GB
L40S
48GB
A100
40GB
A100
80GB
H100 PCIe
80GB
H100 SXM
80GB HBM3
A40
48GB
L4
24GB
V100
32GB
A6000
48GB
RTX 3090
24GB
RTX 6000 Ada
48GB
RTX 4090
24GB
L40S
48GB
A100
40GB
A100
80GB
H100 PCIe
80GB
H100 SXM
80GB HBM3
Trust

Your box.
Your keys.

Isolation is not a feature you toggle on. It is how every rental works, from the first second to teardown.

SSH-only Your key only No exposed backend Per-second metering Hard spend caps Prepaid credits

Your SSH key only

We inject your public key at boot. No passwords, no shared logins, no backdoor.

No exposed backend

The box reverse-tunnels to our gateway. The provider's real IP never reaches you.

Dedicated hardware

You rent the whole GPU. No shared tenancy, no noisy neighbors on your silicon.

Hard spend caps

Set a cap per box. We reclaim it the second the cap is hit. Overruns are impossible by design.

For developers

Built by devs.
For devs.

No console gymnastics. A real machine, your key, one hostname. Bring the workload you already have.

SSH-native

Root access from the first second.

Any framework

PyTorch, TensorFlow, CUDA, your own container.

Per-second billing

The meter starts at connect, stops at teardown.

No lock-in

It is a real box. Bring it, leave anytime.

# your connection string points only at IPX
$ ssh -p 20137 root@gpu.ipxtechholding.com
root@gpu:~# nvidia-smi
# bring any workload, drivers preinstalled
$ pip install torch
$ python train.py --gpus 1
# stop the meter the moment you are done
root@gpu:~# exit
# billing stops at teardown, no minimums
Built for real workloads
01 / 04

"Spin up an eight-GPU node, train for an hour, tear it down. You pay for the hour, not the month."

Training
on dedicated hardware

"Serve a model behind your own SSH tunnel, scaling boxes up and down as traffic moves."

Inference
on dedicated hardware

"Push a render to a 4090, pull the frames, and shut it off before dinner."

Rendering
on dedicated hardware

"Fine-tune on an H100, checkpoint to your bucket, release the box the moment you are done."

Fine-tuning
on dedicated hardware
How you pay

billed by the second

How you pay

no idle minimums

How you pay

from $0.14/hr

How you pay

prepaid and capped

Runs the stack you already ship

PyTorchTensorFlowJAXCUDAvLLMHugging-FaceOllamaComfyUITritonDockerPyTorchTensorFlowJAXCUDAvLLMHugging-FaceOllamaComfyUITritonDocker
Pricing

Honest, metered,
capped.

Prepaid credits, billed by the second. You are never charged past your balance, and metering stops the moment a box is torn down.

Hourly Monthly est.
01

RTX 2080 Ti

22GB VRAM · 1× GPU · United States, US

$0.14 $103.82 /hr/mo est.
  • Dedicated GPU, not shared
  • Root SSH from boot
  • Billed by the second
  • Your SSH key only
  • Tear down anytime
Get started
03

RTX PRO 6000 WS

97GB VRAM · 1× GPU · California, US

$2.54 $1852.58 /hr/mo est.
  • Dedicated GPU, not shared
  • Root SSH from boot
  • Billed by the second
  • Your SSH key only
  • Tear down anytime
Get started

Every box includes root SSH, your key injected at boot, and per-second billing. See live GPUs.

Ready to rent
your first GPU?

From $0.14/hr, billed by the second, on prepaid credits. Launch a real machine in about ninety seconds.

Prepaid credits · no commitment · stop by typing exit