Nirvana Instance Types are live: Compute Families Built for Your Workload

TL;DR:

• 26 standardized instance types across 3 families — n1-highcpu, n1-standard, and n1-highmem
• Each family is optimized for a workload profile: general-purpose, memory-intensive, or compute-heavy
• Up to 37% cheaper than AWS on equivalent configs
• Available now on VMs and NKS

Today we're shipping standardized instance types on Nirvana. 26 types across 3 families, each designed around a specific workload pattern. Pick a family, pick a size, and deploy — on both VMs and Nirvana Kubernetes Service (NKS).

Three Families, Three Workload Profiles

All families follow a consistent naming scheme: {series}-{family}-{vcpus}. Teams coming from other cloud providers will find a familiar structure.

The N1 series runs on AMD EPYC 7513 processors (32 Core / 64 Threads, 3.25 - 3.8 GHz) and is available in three families:

n1-highcpu - 2 GB per vCPU

Built for compute-bound work. CI/CD pipelines, video encoding, cryptographic verification, blockchain validation — workloads where CPU cycles matter more than memory. Sizes range from n1-highcpu-2 (2 vCPU, 4 GB) to n1-highcpu-96 (96 vCPU, 192 GB).

n1-standard - 4 GB per vCPU

The general-purpose family. Web servers, application backends, relational databases, and microservice fleets. This is where most teams start. Sizes range from n1-standard-2 (2 vCPU, 8 GB) to n1-standard-192 (192 vCPU, 768 GB).

n1-highmem - 8 GB per vCPU

Memory-optimized for workloads that need large memory-to-CPU ratios. In-memory caches (Redis, Memcached), analytical databases (ClickHouse, DuckDB), blockchain archive nodes, and data warehousing. Sizes range from n1-highmem-2 (2 vCPU, 16 GB) to n1-highmem-96 (96 vCPU, 768 GB).

Cloud Equivalents and Pricing

The families and memory ratios map to GCP and AWS — if you're migrating from either platform, you already know which family you need.

Pricing is flat: vCPU and RAM at hourly rates. Across equivalent workloads, savings range from ~26% (n1-standard vs. m6a) to ~37% (n1-highcpu and n1-highmem vs. c6a/r6a).

Full pricing breakdowns are in our FAQ and pricing page.

Available Now on VMs and Kubernetes

Instance types are rolling out this week across Nirvana VMs and Nirvana Kubernetes Service (NKS). The same family names, memory ratios, and pricing apply whether you're provisioning individual machines or scaling node pools in Kubernetes.

Log in to your Nirvana dashboard, head to Compute, then VMs and click create. You'll see n1-highcpu, n1-standard, and n1-highmem, listed with their size options. Pick an instance type. Deploy.

If you're running blockchain RPC, indexers, databases, data pipelines, or any latency-sensitive infrastructure - there's a Nirvana instance family built for it.

Nirvana: Modern Cloud for Real-time Workloads

High Clock Speed Compute. Low Latency Storage. Radically Cheaper Bandwidth.Powering Web3, AI, and real-time systems.

Learn more at Nirvana Labs

Nirvana Cloud | Pricing | Blog | Docs | Changelog | Twitter | Telegram | LinkedIn