Latitude.sh

What's good about Latitude.sh

• 25-region coverage with instant deploy options in the US, EU, LATAM, and APAC
• GPU nodes ship with dual 100 Gbps fabric, large RAM, and NVMe for AI training
• API-first platform with detailed REST documentation for automation
• Published hourly pricing for both metal and GPU catalogs

Latitude.sh GPU prices

3 GPU models across 3 configurations. Rank compares Latitude.sh's cheapest on-demand listing for each GPU with every provider's cheapest.

What a run costs on Latitude.sh

24 hours at today's cheapest listing for each GPU, at the size it is listed.

• 1× H100 SXM for 24 hours

  At $1.68/GPU-hr

  $40.32

  $38.40 under the market median rate

• 8× RTX PRO 6000 for 24 hours

  At $3.00/GPU-hr

  $576.00

  $187.64 over the market median rate

• 8× HGX B300 for 24 hours

  At $8.00/GPU-hr

  $1,536.00

  $184.32 under the market median rate

The market median is every provider's cheapest on-demand price for that GPU.

Billing & terms

GPU Bare Metal Hourly

Hourly billing for dedicated GPU servers including RTX 6000 Ada configurations with transparent published pricing.

GPU Virtual Machines

Cost-effective virtualized GPU options with H100 and L40S available in select regions.

CPU Bare Metal Plans

Hourly and monthly billing options for CPU-focused bare metal servers with regional pricing variations.

Transparent Regional Pricing

Published hourly rates for all GPU and CPU plans with pricing that varies by region (US, Brazil, Europe, APAC).

Regions

25 locations across Dallas, Los Angeles, New York, Chicago, Ashburn, Miami, London (2), Frankfurt (2), Amsterdam, Sao Paulo (2), Mexico City, Buenos Aires, Bogota, Santiago, Singapore, Tokyo (2), and Sydney (2).

Support

API reference, contact sales, and a trust center; platform tooling exposes SSH, RAID, and user-data options on provision.

Watch out for

• GPU catalog is narrower than hyperscalers
• Some regions show limited or unavailable GPU stock depending on plan

Features

Automated bare metal

Provision servers with user data, RAID, and SSH over a documented REST API.

GPU Accelerate fleet

RTX 6000 Ada, H100, and L40S capacity with dual 100 Gbps networking on multi-GPU nodes.

Global footprint

25 locations spanning the US, Europe, Latin America, and Asia-Pacific.

Transparent pricing

Hourly and monthly rates published for GPU and CPU bare metal.

Complete infrastructure stack

Beyond compute: managed databases, storage, networking, and Kubernetes services.

Services

• Accelerate GPU Bare Metal

  Dedicated GPU servers tuned for AI training and inference.

• GPU Virtual Machines

  Virtualized GPU plans for quick starts and cost-effective deployment.

• Metal Bare Metal

  General-purpose dedicated servers with high core counts and NVMe.

Getting started

Getting started guide coming soon.