Latitude.sh
A global bare metal cloud with fully automated provisioning across 20 locations, competitive Gen4 AMD hardware pricing, and dedicated GPU clusters for AI workloads, all deployable on-demand without long-term contracts.
Pricing Plans
| Plan | Price/mo | CPU | RAM | Storage | Bandwidth | Notes |
|---|---|---|---|---|---|---|
| m4.metal.small | $296.00 | AMD 4244P (6 cores @ 3.8GHz) | 64GB | 2x960GB NVMe | 20TB free egress | Entry Gen4 bare metal, DDR5, up to 10Gbps |
| f4.metal.medium | $555.00 | AMD 4564P (16 cores @ 4.5GHz) | 128GB | 2x1.9TB NVMe | 20TB free egress | High clock speed, good for latency-sensitive workloads |
| m4.metal.large | $1482.00 | AMD 9254 (24 cores @ 2.9GHz) | 384GB | 2x3.8TB NVMe | 20TB free egress | Mid-range multi-core, up to 5 locations |
| g3.h100.small | $1230.00 | 1x NVIDIA H100 80GB | 80GB VRAM | 2x3.8TB NVMe | 20TB free egress | Single-GPU AI/ML instance, hourly billing available |
· Updated
About Latitude.sh
Latitude.sh runs bare metal servers with the provisioning speed and API-first workflow people expect from cloud VMs, deployed in minutes across 20 locations spanning North America, Europe, Latin America, and Asia-Pacific. It has grown quickly as an alternative to the more established bare metal players for teams that want real hardware without the multi-day provisioning timelines older dedicated providers are known for.
Hardware and Instance Types
The current Gen4 lineup runs DDR5 RAM, NVMe storage, and up to 200 Gbps networking on AMD EPYC and Ryzen processors, spanning small 6-core instances up to 96-core, 1.5TB RAM configurations for heavier workloads. A separate GPU line offers dedicated NVIDIA H100 and RTX Pro 6000 clusters for AI and ML workloads, billed hourly or at discounted monthly and yearly rates.
Pricing
Pricing is transparent and published directly on the site, running from around $296/month for an entry Gen4 server up to several thousand dollars for high-core-count or GPU configurations. On-demand hourly billing is available throughout, with monthly billing running roughly 50% cheaper than hourly and yearly commitments discounted further still, a meaningful saving for predictable long-term workloads.
Deployment and Automation
Everything is API and Terraform-driven, reflecting Latitude.sh’s positioning as bare metal built for teams used to cloud-native infrastructure workflows rather than a traditional dedicated server ordering process. All servers include 20TB of free outbound bandwidth per month with unmetered inbound traffic.
Available in: Dallas, Los Angeles, New York, Chicago, Ashburn, Miami, London, Frankfurt, Amsterdam, Sao Paulo, Mexico City, Buenos Aires, Bogota, Santiago, Singapore, Tokyo, Sydney
Learn More
For more information about Latitude.sh, visit latitude.sh.