News
DealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory鈥檚 NVIDIA & AMD Enterprise ExpansionDealsCorvex agrees USD 103 million take-or-pay deal to perform AI inference with Mid-Atlantic GPUsSovereignNeysa scales AI infrastructure in India with AMD Instinct MI350X GPUs on TytanInferenceMicrosoft Foundry adds Baseten, DeepSeek AI inference (MSFT:NASDAQ)ClustersDataOne and Rowan University sign engineering co-op program agreement for DataOne's Vineland AI factoryClustersLook inside Cal Poly's $3 million AI Factory (Video)DealsOpenAI Taps AMD Chips for 457 MW of AI Capacity in Six-Year TensorWave DealClustersCal Poly launches an AI data center on campus. Take a look insideInferenceArista takes aim at the next AI network battleground: distributed inferenceDealsOpenAI Signs 457 MW AI Data Center Deal as AMD Expands Its Challenge to NvidiaClustersBehind Dell AI Factory鈥檚 NVIDIA & AMD Enterprise Expansion
EveryNeocloud
Oblivus

Oblivus

On-demand and reserved GPU cloud for training, inference and rendering

Oblivus is a GPU cloud provider that rents out graphics compute for machine learning training, inference, rendering and gaming. The company says its platform runs on the latest RTX 5090s and can scale from a single instance to thousands of GPUs, with per-minute billing and no usage quotas.

Share on WhatsApp

  • Offers Nvidia RTX 5090 instances on demand
  • Scales from one GPU to thousands, the company says
  • Per-minute billing with base storage included
  • Claims prices up to 80% below rival clouds
  • API with more than 30 endpoints for automation

What it is

Oblivus sells access to graphics processing units over the internet, a business usually called a neocloud. The company says its fleet includes the latest RTX 5090 cards and that customers can run anything from a single virtual CPU up to tens of thousands of vCPUs, or from one GPU to thousands of them. Workloads it targets include large-scale model training, inference (running a trained model to produce answers), rendering and even remote gaming. Instances are billed by the minute while running, with base storage folded into the price and no setup or termination charges. An OblivusAI operating system ships with common machine learning libraries already installed.

The compute

The company highlights Nvidia RTX 5090 hardware as its newest offering and says its infrastructure suits demanding workloads. Capacity can be taken on demand or held by reservation, and instances can be deployed, resized and scaled through the console or a documented API with more than 30 endpoints. That API supports automated deployment and infrastructure-as-code style management. Oblivus describes its software layer, OblivusAI, as bundling the libraries needed for model work so users do not have to assemble an environment themselves. Details of networking, storage architecture and cluster design are not given.

Customers and deals

Oblivus describes its users as individuals and teams working on machine learning, research, rendering and gaming. Personal accounts suit solo developers and small projects, while business accounts add priority support with an eight-hour response guarantee, wire transfer payments, automated top-ups and volume discounts. Organizations can be created inside the platform, with invitations by email, role-based permissions, controlled billing access and shared instances. The company says it imposes no quotas and runs no verification process before granting access.

Money and ownership

Billing is pay-as-you-go with no long-term contracts, deducted every minute an instance runs and stopped as soon as it is terminated. Reservations offer discounted rates for longer commitments. Payments can be made by card, through the integrated Oblivus Pay system, by bank transfer for business accounts, or via automatic top-ups when a balance falls below a set threshold. The company says it keeps prices up to 80% below competitors by running efficiently and spending little on advertising. Testing can start with as little as $5; there is no free trial, though promotional credits are sometimes offered.

Questions people ask about Oblivus

What does Oblivus sell?

It rents out GPU and CPU capacity by the minute for tasks such as model training, inference, rendering and gaming. Instances can be taken on demand or reserved for longer periods, and the company says it can scale from one GPU to thousands.

Which chips does it use?

The company advertises the latest generation Nvidia RTX 5090 graphics cards as available on demand. No other accelerator models are named on its own pages.

How does pricing work?

Charges accrue per minute while an instance runs and stop the moment it is terminated. Base storage is included, and there are no setup or termination fees. Reservations bring discounted rates, and the company claims savings of up to 80% versus competitors.

Is there a free trial?

No traditional trial is offered. The company says accounts can be tested with as little as $5, and it occasionally issues promotional credits. There are no quotas or verification hurdles before you can deploy.

Can teams use it together?

Yes. Organizations can be created, members invited by email, and roles assigned with different permissions. Owners control billing and resources, while members get access according to their role, and instances can be shared across the team.

Good reading and sources

    Where this page comes from

    There is no English Wikipedia article on Oblivus yet, so the facts on this page come from the sources listed above: oblivus.com/. Some are the subject's own pages, so its claims about itself are reported as its claims. EveryNeocloud wrote the text in its own words, with AI assistance, and compares it with the sources to make sure no passage is copied.

    It may contain mistakes, and figures change over time. Nothing here is investment or buying advice. If something is wrong, tell us. Last refreshed 2026-11-14.

    The weekly digest

    Every Monday: the provider of the day, new pages on EveryNeocloud, and the conferences and anniversaries of the week ahead.