On this page
Today we are launching FlexDedi: single-tenant bare-metal servers that you order and manage like cloud instances. There are ten configurations from $299/mo to $2,999/mo, unmetered traffic on every port and ten locations across North America, Europe and Asia-Pacific.
Why we built it
Teams that run serious workloads usually pick between two options, and both leave something on the table.
Cloud virtual machines deploy in seconds and come with an API, but the hardware is shared through a hypervisor. Performance varies with your neighbors, and outbound traffic is billed per gigabyte, which makes the monthly bill hard to predict for anything that serves a lot of data.
Traditional dedicated servers give you the whole machine at a fixed price. Delivery, however, often takes days, changes go through support tickets, bandwidth is frequently capped, and setup fees are common.
We wanted the useful half of each: the whole machine, ordered and managed the way you manage cloud resources, at a flat monthly price. That is what "dedicated servers, deployed like cloud" means in practice.
The lineup
Every configuration is single-tenant: all of the cores, all of the memory and the full NVMe bandwidth belong to you, with no hypervisor in between. Prices are in USD per month, excluding taxes.
| Server | CPU | Cores / threads | Memory | Storage | Port | Price |
|---|---|---|---|---|---|---|
| Core 16 | AMD EPYC 4545P | 16 / 32 | 128 GB DDR5 ECC | 2 × 1.92 TB NVMe Gen4 | 10 Gbps | $299/mo |
| Core 16X | AMD EPYC 4585PX | 16 / 32 | 192 GB DDR5 ECC | 2 × 3.84 TB NVMe Gen4 | 10 Gbps | $399/mo |
| Storage 264 | AMD EPYC 4565P | 16 / 32 | 128 GB DDR5 ECC | 12 × 22 TB HDD + 2 × 1.92 TB NVMe | 25 Gbps | $699/mo |
| Pro 32 | AMD EPYC 9355P | 32 / 64 | 384 GB DDR5-6000 ECC | 2 × 3.84 TB NVMe Gen5 | 25 Gbps | $749/mo |
| GPU L40S | AMD EPYC 9355P, 1 × NVIDIA L40S 48 GB | 32 / 64 | 256 GB DDR5-6000 ECC | 2 × 3.84 TB NVMe Gen5 | 25 Gbps | $979/mo |
| Pro 48 | AMD EPYC 9455P | 48 / 96 | 576 GB DDR5-6000 ECC | 2 × 7.68 TB NVMe Gen5 | 25 Gbps | $999/mo |
| Pro 64 | AMD EPYC 9555P | 64 / 128 | 768 GB DDR5-6000 ECC | 4 × 3.84 TB NVMe Gen5 | 2 × 25 Gbps | $1,349/mo |
| Max 96 | AMD EPYC 9655P | 96 / 192 | 1,152 GB DDR5-6000 ECC | 4 × 7.68 TB NVMe Gen5 | 2 × 25 Gbps | $1,899/mo |
| GPU RTX PRO | AMD EPYC 9455P, 2 × NVIDIA RTX PRO 6000 Blackwell 96 GB | 48 / 96 | 576 GB DDR5-6000 ECC | 2 × 7.68 TB NVMe Gen5 | 25 Gbps | $2,199/mo |
| Max 192 | 2 × AMD EPYC 9655 | 192 / 384 | 1.5 TB DDR5-6000 ECC | 4 × 7.68 TB NVMe Gen5 | 2 × 100 Gbps | $2,999/mo |
A few notes on how the range is built:
- Core uses AMD EPYC 4005 processors with boost clocks up to 5.4 GHz, or 5.7 GHz on the Core 16X, whose 3D V-Cache brings 128 MB of L3. It suits web apps, game servers and CI runners that reward fast single threads.
- Pro and Max use AMD EPYC 9005 processors with NVMe Gen5 storage. Memory sizes follow the 12 DDR5 channels of each socket, which is why you see 384, 576, 768 and 1,152 GB rather than round numbers.
- GPU pairs EPYC 9005 hosts with an NVIDIA L40S or two NVIDIA RTX PRO 6000 Blackwell cards, for inference, rendering and fine-tuning.
- Storage 264 puts 264 TB of raw disk behind a 25 Gbps port, for backups, object storage and media libraries.
Configurations marked "Ready in minutes" are pre-racked and burn-in tested, and are usually delivered within 15 minutes of payment with the operating system installed. Storage, GPU and Max configurations are assembled to order and usually delivered the same day. The full specifications are on the dedicated servers page.
Unmetered traffic, on every port
Every server includes unmetered traffic: no monthly cap, and no egress or overage fees, inbound or outbound. The port is the limit, at 10, 25, 2 × 25 or 2 × 100 Gbps depending on the server.
We chose this model because per-gigabyte billing punishes exactly the workloads that dedicated servers are good at, such as streaming, downloads, backups and busy APIs. With a flat price, a traffic spike or a viral week changes your load, not your invoice. We explain the trade-offs between billing models in Unmetered bandwidth explained.
Ten locations on three continents
FlexDedi servers run in New York, Ashburn, Dallas and Los Angeles in North America; Amsterdam, Frankfurt and London in Europe; and Singapore, Tokyo and Sydney in Asia-Pacific. All are carrier-neutral data centers with redundant power and cooling, connected by a backbone with diverse transit and peering.
The Core series is available in all ten locations, and each other configuration is listed with the locations that offer it.
What every server includes
The monthly price covers more than the hardware:
- always-on DDoS protection on every IP address
- IPMI and an HTML5 KVM console, so you can reach the server even when its operating system is down
- one IPv4 address and a /64 IPv6 block, with additional IPv4 addresses on request
- free private networking (VLANs) between your servers in the same location, with internal traffic not billed
- one-click images for Ubuntu, Debian, AlmaLinux, Rocky Linux, Windows Server and Proxmox VE, plus any custom ISO through the KVM console
- a 99.99% monthly network uptime SLA with service credits
- 24/7/365 support by engineers, by ticket, with spare parts on site in every location for fast hardware replacement
There are no setup fees and no long-term contract. Billing is monthly in advance, and if you cancel, your server stays active until the end of the period you paid for.
Servers are unmanaged by default: you get full root or administrator access, and you remain responsible for your operating system, software and backups. If you are moving from another provider, our engineers will help you plan the migration at no charge. Custom builds and fleet pricing from five servers are available by quote.
What comes next
Next on our list is the rollout of the client area, the REST API and the Terraform provider.
The control panel is built around the day-to-day actions: reinstall, reboot, rescue mode, reverse DNS and the KVM console. The REST API is designed to cover every one of those actions, so anything you can click, you can script. The Terraform provider builds on the API, so you can declare servers next to the rest of your infrastructure code and review changes before you apply them.
We will document each step in the changelog as it reaches customers.
Getting started
Pick a configuration and a location, choose an image and add your SSH key. If you are unsure which server fits your workload, our engineers will help you size it.
Thank you for reading, and welcome to FlexDedi.