You can now access the MAC addresses of server network interfaces through both the API and Terraform provider.

The updated API now returns interface names and MAC addresses to enable easier cluster bootstrapping, and includes rack ID for data center-level redundancy management.

Learn more about it in our API reference Docs.

The newest version of Custom Images is now available in Tech Preview.

This update enables you to create, manage, and deploy custom OS images more easily across your Latitude.sh devices using iPXE scripts.

We have completely revamped how Custom Images work to enable more flexible server provisioning. Deployment reliability has been enhanced, and the average boot time has been reduced from 20 minutes to ~3 minutes. The new version also includes DHCP support, so you no longer need to set static IP addresses in your iPXE scripts.

The new Custom Images is currently available in Ashburn, with other locations in the US East region to follow. To join the Tech Preview, fill out this form.

For more information on how to use this feature, check out the Custom Images section in our docs.

Version 2.0.0 of our Terraform provider is now available with a complete architectural upgrade that brings significant improvements while maintaining compatibility for most configurations:

• Performance: 30-50% faster operations when compared to the previous version, optimized API calls and state management, reduced memory footprint, and better resource efficiency.

• Reliability: Runtime error prevention with compile-time type checking, improved state management that better handles edge cases, and enhanced error messages to help you fix issues.

• Future-proof architecture: Built on Hashicorp's latest framework, ready for the upcoming Terraform 2.0, and with modern patterns that speed up the addition of new features.

While most configurations will function normally, some important changes have been introduced due to the addition of new API endpoints. You can check out the Terraform Docs for an in-depth view of what's changed and our Migration Guide to help you get started with the version upgrade.

With 24 high-speed cores and massive memory allocation, the f4.metal.large is designed for latency-sensitive applications that demand exceptional performance per core.

Here's the complete spec:

• CPU: AMD 9275F (24c @ 4.1 GHz)

• RAM: 768 GB of DDR5

• Storage: 2x 480 GB + 2x 3.8 TB NVMe

• Network: Dual 100 Gbps NICs

Starting at $1.52 per hour, the f4.metal.large is now available in Ashburn, Chicago, and Dallas.

More locations are coming soon, and you can now reserve this instance for Miami, Los Angeles, New York, London, Frankfurt, Singapore, and Tokyo.

We’ve launched a brand-new interface for the homepage in our dashboard, designed to give you a faster and more straightforward overview of everything happening across your infrastructure. What's new:

• At-a-glance resource counts with respective locations

• Cost, usage, and bandwidth insights, filterable by region

• Quick access to Learn, Support, and Changelog

Check it out