With high clock speed and generous memory, the f4.metal.small is ideal for latency-sensitive applications that need fast and reliable compute performance without overprovisioning.

Here's the complete spec:

• CPU: AMD 4484PX (12c @ 4.4GHz)
• RAM: 96GB of DDR5
• Storage: 2x 960 GB NVMe
• Network: Dual 10 Gbps NICs

Starting at $0.40 per hour, the f4.metal.small is now available in Ashburn and New York.

More locations are coming soon, and you can now reserve this instance for Los Angeles, Frankfurt, and Tokyo.

The new and improved Remote Access is now available for instances deployed in all sites, enabling access to your server's IPMI without requiring a VPN and generating credentials with a single click.

Try it out

We’ve expanded our network options for crypto payments. The Ethereum network has been added, and you can now purchase credits using tokens through either the Solana or Ethereum networks.

When adding credits, select your preferred network, choose your desired token, and sync the respective wallet.

Visit the Billing page to get started.

We've improved the Usage page interface to give you more control, better insights, and a faster experience:

• Cleaner layout: A more intuitive structure that makes key information easier to find
• More granular data: Dive deeper into your usage metrics
• Improved speed: Faster load times for a smoother experience

You can switch to the previous layout anytime while the new page is in beta. We’d love to hear your feedback.

Our official Python SDK is now live. From provisioning instances to managing firewalls and IPs, the SDK wraps our API in a clean, Pythonic interface with full support for type hints, error handling, and pagination. It includes:

• Easy setup with pip or poetry, and built-in API key authentication
• Full resource coverage for projects, facilities, instances, firewalls, IPs, and more
• Great DX with type safety, pagination support, and IDE autocompletion

Check out the SDK docs to learn more.