Best VPS by use case

Pick your use case. We'll show you the VPS providers we'd actually run it on — tested, not just listed.

Running more than one app? Build your stack → for a sizing estimate across your whole setup.

25 use cases

Best VPS forAI & LLM Apps

Serving an open LLM — Ollama, LocalAI, or Open WebUI — is a RAM story first and a CPU story second. The model weights have to live in memory to run at all, so the winning box is the one that gives you the most GB of RAM and fast NVMe per dollar. Every pick here clears the 8 GB floor where small models become usable.

See 5 picks →

Best VPS fora CRM

Self-hosted CRMs — Twenty, EspoCRM, or a full ERP like Odoo — are the middleweight of self-hosting: an app, a database, and a small team's worth of users. You want a box with enough RAM for the database plus the app, enough disk for attachments, and the reliability to keep your pipeline visible.

See 4 picks →

Best VPS forDatabases (Postgres & MySQL)

A self-hosted database wants three things: RAM to cache the working set, fast NVMe for write latency, and reliable backups. These plans deliver all three without enterprise pricing.

See 5 picks →

Best VPS forDocker

Docker rewards RAM and fast disk far more than raw clock speed. You want a clean image where Docker installs in a minute, full virtualization so containers behave, and enough memory for a real Compose stack.

See 5 picks →

Best VPS forGitea

Gitea is a single Go binary that idles in a couple of hundred megabytes — the server choice is almost never about CPU or RAM, it's about disk. Repositories, LFS objects, container images and CI artifacts accumulate quietly, and the day you turn on Actions runners the box stops being small.

See 4 picks →

Best VPS forGitLab

GitLab CE is not a small install: it bundles PostgreSQL, Redis, Gitaly, Puma, Sidekiq and NGINX behind one Omnibus package, and upstream's own single-node requirements reflect that — 16 GB RAM and 8 vCPU as the baseline, with 8 GB workable only in a memory-constrained setup. That makes GitLab the one workload on this site where recommending a genuinely large VPS is the honest call, not an upsell: the picks below are all providers whose 20 GB+ tiers actually clear GitLab's documented footprint, not their cheapest boxes.

See 3 picks →

Best VPS forHome Assistant

Home Assistant is one of the few self-hosted workloads where stability and small steady load matter more than raw power — it hums along at a fraction of your VM, then spikes when automations fire or the dashboard refreshes. The picks below are the reliable, low-touch boxes people actually run it on.

See 4 picks →

Best VPS forImmich

Immich backs up a lifetime of photos and runs machine-learning search and face detection over them — so it wants real RAM, fast NVMe for the library and thumbnails, and generous bandwidth for the initial mobile upload. The goal is a box that ingests tens of thousands of photos without the ML jobs starving everything else.

See 5 picks →

Best VPS forJellyfin

Jellyfin's sizing question is really two questions: will the server transcode, and how much video will it push out. Direct play to a modern client barely touches the CPU, but one incompatible client forces a software transcode that will saturate every core on a small VPS — and every stream is sustained outbound bandwidth, hour after hour.

See 4 picks →

Showing 1–9 of 25

Showing 1–9 of 25

We earn commissions when you shop through the links below. Full disclosure →

Where to host itaffiliate disclosure

Kamaterafree trial

From $4/mo · 1 vCPU / 1 GB / 20 GB · US + EU + Asia

Start free on Kamatera →

DigitalOceanalso works on

From $6/mo · 1 vCPU / 1 GB / 25 GB · US + EU + Asia

Deploy on DigitalOcean →

Paid link — we earn a commission if you shop through it.

We use analytics cookies (Google Analytics, PostHog) to see which guides are useful. No ad networks, no cross-site tracking. See our privacy policy.