EQVPS

VPS for Ollama and Local LLMs

Self-host small language models on a CPU server — private, unmetered, paid in crypto. Honest about what CPU inference can and can't do. From $12/mo.

Let's get the disappointing part out of the way, because the internet is full of pages that don't.

Our servers are CPU-only. We don't offer GPUs. That means local LLM work here has a hard ceiling, and pretending otherwise would just waste your money.

What actually works on CPU

On the standard plans, with up to 6 vCPU and 6 GB of RAM (our Medium plan — $12/mo), you're in the range of small quantized models:

  • 1B–3B models (Q4): genuinely usable. Fast enough for classification, tagging, routing, and simple structured extraction.
  • 7B–8B models (Q4): run, but expect a few tokens per second. Fine for a queue that chews through documents overnight. Painful as a chat window.
  • Embedding models: excellent fit. They're small, fast on CPU, and this is probably the single best reason to run Ollama on a box like ours.
  • 13B and up: don't. You'll be swapping and waiting.

If you need a fast, interactive 70B chat, you need a GPU host — that's a different product from a different provider, and we'd rather tell you than take your $12.

Where a CPU box genuinely wins

Privacy. Your documents, your prompts, your embeddings — never leaving a machine you control, never becoming someone's training data. For a lot of people that's the entire point, and a few tokens per second is an acceptable trade.

Cost predictability. No per-token bill. A pipeline that classifies fifty thousand records costs the same as one that classifies fifty: $12.

Async work. Most useful LLM work isn't a chat window. It's a queue: summarize these, tag those, embed this corpus. A CPU box grinding through a backlog overnight is perfectly good at that.

Setup

# Ubuntu 24.04 — Medium plan recommended
curl -fsSL https://ollama.com/install.sh | sh

# Start with something small and see for yourself
ollama pull llama3.2:3b
ollama run llama3.2:3b "Summarize this in one sentence: ..."

# Embeddings — the sweet spot for CPU
ollama pull nomic-embed-text
curl http://localhost:11434/api/embeddings \
  -d '{"model":"nomic-embed-text","prompt":"hello world"}'

Ollama serves an HTTP API on localhost:11434. Keep it there. If you need to reach it from elsewhere, put it behind a reverse proxy with authentication on a dedicated-IP plan — never expose an unauthenticated model endpoint to the open internet.

Pair it with a vector database (Qdrant or pgvector in Docker) and you have a complete private RAG stack on one $12 box. That combination — small local embeddings, local vector store, external API only for the heavy generation step — is the setup that actually makes sense on hardware like this.

Picking a plan

UsePlanPrice
Embeddings, 1–3B models, RAG pipelineMedium$12/mo
Same, plus a public endpoint behind authMedium-IP$20/mo
Just testing small modelsSmall$8/mo

CPU-only: standard plans go up to 6 vCPU and 6 GB RAM, the Pro line up to 12 vCPU and 80 GB. NVMe storage, unmetered traffic, Germany or Finland. Crypto payment, no KYC. Yearly billing is one transfer instead of twelve.


Ready? Deploy a server → — live in about a minute, paid in crypto, no ID required.

VPS for Open WebUI: a private chat interface for LLMs →

Ready to deploy? Pay with crypto, no KYC — live in about a minute.

Deploy now →

FAQ

Can I run Llama 70B on this?

Not in a useful way. Our standard plans are CPU-only with up to 6 GB of RAM — the range of small quantized models (roughly 1B–8B at 4-bit). The Pro line goes up to 80 GB of RAM and 12 vCPU, enough to load a 4-bit 70B model, but on a CPU it answers very slowly. For real work with 70B you need a GPU. We don't offer GPUs, and we'd rather say so than sell you a disappointment.

How fast is CPU inference, really?

Expect a few tokens per second on a 7–8B model at 4-bit quantization, and noticeably better on 1–3B models. That's fine for background jobs, embeddings, classification, and async pipelines. It is not fine for a snappy interactive chat.

So what is this actually good for?

Private embeddings, classification, summarization queues, RAG pipelines where latency doesn't matter, and keeping data off third-party APIs. Also: learning, testing, and prototyping before you rent a GPU somewhere.

Do you ask for ID?

No. Email to register, USDC or USDT to pay. Which is often the reason people want a private model in the first place.

Comments

No comments yet. Be the first.

Leave a comment

Comments are moderated before they appear.