Starter
Lab / Dev
Usage-based or fixed lab allowance
Pro
Teams
Volume and commitment discounts
Feature comparison
Capabilities by tier. InferoFabric is a software platform; GPU capacity is provided by your infrastructure or our managed service.
| Feature | Starter | Pro | Enterprise |
|---|---|---|---|
| Inference endpoints & autoscaling | |||
| Training jobs & checkpoints | |||
| Cost caps & auto-suspend | |||
| Verified stack blueprints | |||
| Hybrid placement (on-prem + cloud) | |||
| Projects & environments | Up to 2 | Unlimited | Unlimited |
| Usage & cost attribution | |||
| API & CLI access | |||
| Documentation & community support | |||
| SLA | |||
| Email support | |||
| Priority / dedicated support | |||
| SSO & advanced RBAC | Optional | ||
| Custom deployment & air-gap | |||
| MSP / multi-tenant options |
Deployment options
InferoFabric is a software platform that orchestrates GPU workloads. You choose how and where to run it.
Software platform
InferoFabric orchestrates workloads on supported infrastructure. You bring your own Kubernetes clusters (on-prem or in the cloud) and run the InferoFabric agent; the control plane handles placement, policies, and cost control. We do not resell raw GPU capacity—we provide the platform that runs on infrastructure you control or that is powered by major cloud providers.
Optional managed service
For teams that prefer not to operate the platform themselves, we offer an optional managed service: we run and operate InferoFabric (control plane and, where chosen, worker capacity) for you. GPU capacity in a managed deployment is powered by major cloud providers; InferoFabric orchestrates workloads on that supported infrastructure. Pricing and terms are provided on request.
Quick ROI estimate
Deployment and pricing FAQ
How you can run InferoFabric and what to expect from pricing.