Sign up and launch,
rent GPUs by the minute
Glows.ai is KONST's GPU cloud platform partner, and its compute comes from data centers that KONST built or operates under contract. Register online and choose an instance type to launch an instance. The operating system, drivers, and frameworks are already set up, so you do not need to build your own data center or environment. Instances are billed by the minute, billing stops when you shut down, and there is no monthly fee or minimum usage.
Set up GPUs by use: development, team sharing, online inference
On-demand cloud launches single-machine or multi-GPU instances, suited to development and short training runs. A virtualized cluster pools nodes into a resource pool that a team shares. Inference service comes in shared and dedicated types, for latency-sensitive online services.
Interactive development and short training runs
For experiment environments you start and stop at will: launch single-machine or multi-GPU instances on demand, and run as soon as you choose an instance type.
Long training runs shared by a team
When several people compete for the same GPUs, multiple nodes are combined into a schedulable resource pool that the team allocates flexibly.
Online inference
For latency-sensitive traffic that cannot be interrupted: the shared type is billed by the minute, and the dedicated type reserves fixed resources.
Launch and start training right away
No need to build a data center or environment. Choose an instance type and run. Instances are billed by the minute, and billing stops at shutdown.
Ready on launch
The operating system, drivers, and frameworks are already set up. Sign up, choose an instance type, and start running.
Per-minute billing
No monthly fee, no minimum usage
No rebuild when changing instance types
Data is mounted at launch and kept after shutdown, and a snapshot can be restored to a new instance.
Data persists, and changing instance types needs no environment rebuild
Data and environments do not disappear with the instance lifecycle. After changing instance types, an interruption, or a restart, you can continue from the previous state. Models and private data are stored separately.
Data persists after shutdown
Data is mounted at launch and kept after shutdown, so you do not need to upload it again each time.
Environment snapshot save and restore
A snapshot saves the full state of the environment, so you can continue directly after changing instance types or an interruption.
Public access · Private storage
Public models and datasets are available directly. Private data is stored in your personal account space, and a high-performance storage service is available for large volumes.
Self-service online launch, billed by actual usage
Register online to choose an instance type and launch an instance yourself, with no contract required. Billing is by the minute and stops at shutdown, with no monthly fee or minimum usage. You top up first and usage is deducted from the balance, and spending can be checked in real time.
Switch to a high-end GPU and
training time falls from 8.5 hours to 70 minutes
The unit price is higher, but run time drops sharply, so total spend under per-minute billing falls and the time saved goes straight into the next round of experiments.
Centralized team top-ups, with credits allocated by role
One account manages multiple members. Credits are topped up in one place and then allocated, so members do not each have to pay. Administrators can restrict which instance types and images members can use, so credits are not used on high-end GPUs.
PermissionsCreator and owner of the team, with all permissions
PermissionsActivate accounts, allocate credits, control resource permissions, release instances
PermissionsLaunch and use instances within the assigned credits and resource permissions
FAQ
How are Glows.ai GPU cloud fees calculated?
Are there special plans for schools or research institutions?
Do I have to rebuild my environment if I change instance types?
Which GPU should I use for small models and for large models?
Services often evaluated together
Ready to launch your first GPU instance?
Tell us your team size and use case, and we will reply with the suitable plan and credit allocation