Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.hyperbolic.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

Getting Started with Reserved Clusters

Work with our specialists to design your perfect cluster configuration and get deployed in as little as 3-5 days.

The Reservation Process

1

Initial Consultation

Timeline: Same day
  • Schedule a call with our Head of Sales
  • Discuss your requirements and use cases
  • Receive a custom configuration proposal
  • Get transparent pricing with all discounts applied

Schedule a Call

Book a consultation with our sales team
2

Configuration Design

Timeline: 1-2 days
  • Work with our engineering team on specifications
  • Review network topology and storage options
  • Finalize hardware selection
  • Approve the deployment plan
See Cluster Configuration for available options.
3

Deployment

Timeline: 3-5 days
  • Hardware provisioning and setup
  • Network configuration and testing
  • Software stack installation
  • Security and access configuration
  • Comprehensive benchmarking and validation to ensure performance meets expectations defined during configuration design
4

Go Live

Timeline: Day 5
  • Receive a comprehensive handoff document with cluster details, access information, and configuration specs
  • Receive cluster access credentials
  • Join dedicated Slack support channel
  • Complete onboarding with engineering team
  • Start running your workloads

Contact Sales

Schedule a Consultation

Ready to discuss your needs and get a custom quote?

Quick Questions?

Not ready for a full consultation? Get quick answers:

What to Prepare

Before your consultation, it helps to have answers to these questions:
Understanding your use case (training, inference, research) helps us recommend the right GPU types and configurations.
Minimum reservation is typically 8 GPUs. Larger commitments unlock better volume discounts.
Standard deployment is 3-5 days. Let us know if you have urgent requirements.
For distributed training, InfiniBand or RoCE networking may be essential. For inference, standard networking often suffices.
Longer commitments (3-12 months) unlock additional discounts on top of volume pricing.

Next Steps

View Configuration Options

Explore GPU, networking, and storage specifications