Katapult.io Nvidia H100 GPU, £ (on request) on Cloud

Nvidia H100 GPU has been added on Sep 17, 2025

🔌 Hosted domains :	unlimited
🆓 free domains :	0
📌 Dedicated IPs :	0
💳 Payment Methods :	Credit / Debit / Prepaid Cards
🔨 Control Panel :	KVM
🔧 Category :	Self Managed
✍️ Support Options :	EmailHelp Desk
⚑ Targeting :	UK

See also initial Nvidia H100 GPU plan location on their website!

*📜 Plan description

📄 Editorial Review

Katapult.io positions itself as a high-performance public cloud provider with a strong emphasis on sustainability. Founded with a mission to deliver reliable compute power while reducing environmental impact, Katapult is one of the few B Corp certified cloud companies, powering its infrastructure entirely with renewable energy from solar, wind, and sea sources.

Cloud Infrastructure and Virtual Machines

At the heart of Katapult’s offering are virtual machines powered by AMD EPYC processors, designed for workloads that demand both efficiency and high performance. Customers can select from CPU-optimized instances ("Rocks") for compute-heavy applications, or memory-optimized instances ("Boulders") for real-time data and in-memory processes, scaling up to 96 vCores and 960GB RAM. All VMs are supported by all-flash distributed storage with triple replication, ensuring resilience and fast disk access.

For demanding use cases like AI, rendering, or machine learning, Katapult provides NVIDIA GPU options that can be attached flexibly to virtual machines. These GPU-based instances are tailored for deep learning, inference, and high-performance computing, giving customers scalable GPU power without the need for on-premises hardware.

Storage Solutions

Katapult’s storage services are built on flash-first architecture for maximum throughput and reliability.

• Block Storage for persistent, high-performance volumes
• Elastic File Storage (KEFS), a disaggregated NVMe-over-fabric system offering fault-tolerant and highly scalable file storage
• Object Storage, compatible with S3 API tools, suitable for backups, datasets, and application integration

All options are designed with redundancy, speed, and scalability in mind.

Networking and Security

Katapult’s network stack is based on Cumulus software and hardware from Mellanox and Juniper, ensuring scalability and fault tolerance. Features include:

• Private SDN Networks with free internal data transfer
• Load Balancers with automatic HTTPS redirection, SSL management, health checks, and multiple routing algorithms
• Inclusive DDoS Protection with hybrid Corero appliances and upstream scrubbing capable of mitigating attacks beyond 1Tbps
• Firewall tools that are hypervisor-based and API-manageable, included at no extra cost

Data Center Locations

Customers can deploy resources across multiple regions: London (UK), Amsterdam (NL), Phoenix (US), Edison (US), with Singapore and Tokyo listed as upcoming. While Katapult does not claim ownership of the physical facilities, these locations form part of their accessible cloud regions. All centers are powered by renewable energy, reflecting the company’s sustainability-first approach.

Support and Management

Account management is handled via the Katapult dashboard and a RESTful API, supporting automation and integration. OS templates for Ubuntu, Debian, CentOS, and Fedora allow for quick provisioning.
Support is available via email (team at katapult.io) and ticket-based contact, but no live chat or phone support is currently advertised.

Pricing, Payments, and Guarantees

Katapult offers pay-as-you-go pricing with billing handled through Stripe and PayPal. A £100 free credit is available for new customers to test the platform.

• Uptime Guarantee: SLA of 99.99% per VM. Service credits equal to the downtime are applied to future invoices.
• Refund Policy: No general money-back guarantee is published; refunds are not a standard part of their policy.

Strengths

• 100% renewable-powered infrastructure, B Corp certified
• High-performance AMD EPYC CPUs with scalable GPU integration
• Distributed flash storage with triple replication for reliability
• Strong DDoS protection and enterprise-grade networking stack
• Multiple data center regions with expansion plans
• SLA-backed 99.99% uptime

Weaknesses

• No advertised money-back guarantee
• Limited direct support options (no phone or live chat)
• Still expanding global presence compared to larger cloud providers