AIHostBox
Client area Deploy an app
All 57 applications

Categories

AI assistants & chatLLM runtimesAgents & RAGGateways & LLM opsVector & dataAutomationSpeech, image & videoSearch & web agentsSupport & CXML platformMCP servers

Solutions

Private team chatUnlimited automationCompany knowledge AIAI gateway & observability

More

PricingPrivate LLM hosting LicensingSecurity SupportClient area
57 open-source AI apps

Self-hosted AI apps, managed for you.

Flat monthly pricing on a dedicated single-tenant instance. Your API keys, your data, your logs — installed, patched, monitored and backed up by us. No GPU contracts, no per-seat tax, no token meter.

Browse the catalogue See what you can build
$5.99entry plan / month
57apps in the catalogue
1:1single-tenant instance
24hcritical patch target
aihostbox · deploy / open-webui
00:02pulling open-webui:latest …
00:19attaching volume 8 GB NVMe
00:33issuing TLS certificate
00:41binding chat.yourteam.com
00:52ready → https://chat.yourteam.com
nightly backups managed patching single tenant
Most deployed

Six that cover most of what people come here for

An always-on assistant, a team chat interface, a private model server, an app platform, a gateway and an automation engine. The other 51 are one click away.

OpenClaw logo

OpenClaw

AI assistants & chat

Your always-on personal AI assistant, running 24/7 on a server you control.

$9.99/moDetails →Open WebUI logo

Open WebUI

AI assistants & chat

A ChatGPT-style workspace for your team, pointed at any model you like.

$5.99/moDetails →Ollama logo

Ollama

LLM runtimes

Your private LLM server — pull a model, get an API, keep every token on your own box.

$16.99/moDetails →Dify logo

Dify

Agents & RAG

A visual studio for building LLM apps, agents and RAG pipelines.

$18.99/moDetails →LiteLLM logo

LiteLLM

Gateways & LLM ops

One OpenAI-compatible endpoint in front of every model provider you use.

$5.99/moDetails →Activepieces logo

Activepieces

Automation

Open automation with hundreds of integrations — and no execution quota.

$5.99/moDetails →

See all 57 applications →

Start from the outcome

Or start from what you are trying to build

Four stacks, each shipped as one dedicated instance with its database, cache and backups already wired.

Private team chat

A ChatGPT-style workspace on your domain, with your keys and your data.

Open WebUILibreChatLiteLLMfrom $7.99 / mo

Unlimited automation

Workflow automation with no execution quota — the price does not move when the work does.

ActivepiecesWindmillNode-REDKestrafrom $9.99 / mo

Company knowledge AI

Your documents, answered with citations — and never indexed off your instance.

AnythingLLMDifyOnyxQdrantfrom $16.99 / mo

AI gateway & observability

One endpoint in front of every provider — with budgets, keys and traces you own.

LiteLLMLangfuseOpikfrom $11.99 / mo
What managed actually means

The part nobody else puts in writing

Anyone can start a container. The work is keeping it correct six months later — so here are the numbers we hold ourselves to.

24h

Critical patch window

Critical advisories patched within a day of a fixed release, high severity within three. Agent tools carry minimum reviewed versions.

1:1

One customer per instance

Never a shared workspace. Your own storage, your own admin account, your own network policy — and a licence requirement for several of these apps.

0%

Markup on model tokens

You bring your own provider key and pay them directly. We host the software; we do not sit in the middle of your inference bill.

57

Applications, licence-checked

Every entry vetted against its licence before it went on sale. The ones that failed are documented rather than quietly sold.

Why self-host

The same software, without the SaaS multiple

Official clouds charge for seats, workflow executions or workspace tiers. You pay for a box that runs the software.

OpenClaw Cloud
$49/mo
Vendor-hosted, official
Dify Professional
$49/mo
Billed annually, 50 apps
Managed endpoints
$43/mo+
From $0.06 per hour, always on
AIHostBox
$9.99/mo
Dedicated instance, unmetered runs

Competitor prices verified September 2026. Monthly equivalents assume 730 hours of continuous runtime.

One panel

Everything you run, in one list

Instances, memory headroom, version state and the timestamp of the last backup. No Docker logs to read, no compose file to edit — and if something needs a human, a human is already looking at it.

  • Resize memory and storage in place; your data does not move.
  • Restore any instance from the previous night in one request.
  • Version state visible per app, so "is it patched" is never a question.
How we keep it patched
How it works

Three steps, no Docker knowledge required

Pick the app and the size

Choose memory and storage from the plan table. Every plan is a single-tenant instance with its own volume, its own configuration and its own admin account.

We deploy and harden it

TLS on your domain or ours, firewall, a version pinned to a reviewed release, nightly off-box backups and isolated secrets.

You log in and build

The admin account is yours. Add your API keys, invite your team, export your data whenever you want. Patching stays with us.

Questions

Straight answers

What is managed self-hosted AI hosting?
You get a dedicated instance of an open-source AI application — Open WebUI, Dify, Ollama and so on — running on our infrastructure under your own domain. We install it, keep it patched, monitor it and back it up; the data, the admin account and the API keys stay yours. It is neither a shared SaaS workspace nor a bare VPS you have to administer.
Do I need a GPU to run these apps?
Most of the catalogue never touches one. Chat interfaces, automation tools, RAG front-ends, vector databases and gateways are CPU workloads that call whatever model provider you configure. GPU capacity only matters if you want fast local inference — and for that we publish honest CPU throughput figures per plan rather than hiding them.
Can I use my own OpenAI or Anthropic key?
Yes, and that is the default. Bring your own key: token spend goes to your provider account at cost and we never proxy or mark up model usage. If you would rather run a local model, choose an Ollama or LocalAI plan sized for the model you want.
What happens to my data if I leave?
You can export application data and database dumps at any time, and we hand over the final backup archive on request. These are upstream open-source applications with upstream export paths — nothing is locked in a proprietary format.
Why is n8n documented but not for sale?
n8n ships under the Sustainable Use License, which restricts offering the software to others for a fee. Until we hold written commercial permission we will not sell it. The guide stays because it is useful, and we point at Activepieces, Windmill, Node-RED and Kestra, which are licensed for managed hosting.
How fast do security patches land?
Critical advisories within 24 hours, high severity within 72. Internet-facing agent tools are pinned to a minimum reviewed release — Flowise, for example, is never deployed below 3.0.6 because earlier versions carry a remote code execution advisory.

Put your AI stack on your own box

Pick an app, pick a size, and have it running today. Month to month, cancel whenever.

Browse the catalogue Talk to a human