You built it.
Now launch it.
Deploy from GitHub to your own private server — with fixed compute costs and one operations layer around it. Add security when you need it. Without becoming your own DevOps team.
- Prepare server — Complete
- Pull repository — Complete
- Build image — Complete
- Start container — Complete
- Health check — Complete
- HTTPS certificate — Complete
The demo was the easy part.
AI made software dramatically easier to build. It didn’t remove what production needs. The moment your app works, it quietly starts depending on all of this:
Your app was simple.
Your stack isn’t.
Each of these tools solves its part well. The problem is what happens when one small app slowly turns into a stack of separate services, bills and dashboards.
A common setup after building with AI
Illustrative — not everyone uses exactly these tools.
- ChatGPT / Claudeidea & code
- Cursor / Lovable / Replitbuild
- GitHubsource
- Hosting platformbill + dashboard
- Database / storagebill + dashboard
- Monitoringbill + dashboard
- Security toolingbill + dashboard
- DNS providerdashboard
- Backupsbill + setup
- More SaaS subscriptionsbills
The hidden cost isn’t only subscriptions:
The VIBR way
Keep building with anything. Run it in one place.
You don’t need seven tools to keep one app alive.
Why infrastructure bills start to grow.
Before you compare prices, compare pricing models. One grows with your usage. The other doesn’t.
Model A · Fragmented, usage-based
- Platform feemonthly
- Computeper usage
- Databaseper plan / usage
- Monitoringper plan
- Securityper seat / plan
- Backupsper GB
- Additional servicesvaries
More usage can mean more infrastructure cost.
Model B · Private infrastructure
- Your serverone fixed monthly price
- Control layeroptional, fixed
- Securityoptional, fixed
- Your app’s AIbring your own keys
Your workload runs on infrastructure you control, at a price you know in advance.
Your VIBR cost: exactly the same as last month. The same on a quiet day as on your best day — the prices are at the bottom of this page.
A big spike can make a fixed server slower — it can’t make your invoice bigger. For steady workloads, fixed-price private compute is often the more economical model.
One private layer between your code and production.
Your build tools stay exactly as they are. VIBR Control turns your code into a running, monitored app on a server that belongs to you — with security around all of it.
Your traffic goes straight to your server. VIBR is never in the path, so your apps keep running even if VIBR’s dashboard doesn’t.
Connect. Deploy. Keep it running.
Not: learn Linux.
Connect
Connect your GitHub repository and a server: choose a private VIBR server, or bring a compatible VPS in the current pilot.
Deploy
Pick the branch, add your environment variables and your domain. VIBR builds the app and switches traffic only when it’s healthy.
Keep it running
VIBR watches health, HTTPS, the server and every deploy. If your app stops responding, VIBR finds out why and tells you what to do.
Name your app.
Watch it go live.
This is what happens when you press Deploy. Type your app’s name, pick what it is, and watch VIBR take it from a repository to a live URL — without a single command.
- Prepare server
- Pull repository
- Build image
- Start next to current release
- Health check
- HTTPS certificate
- $ waiting for launch
Is everything running? Is anything wrong? Do I need to act?
The VIBR dashboard answers those three questions in one look — for every app, automation and server you run. Click through it:
Overview
Running — 2 things to look at
Nothing is down. Doctor has the next step for each.
- Deployment failed during health checkRedeployapi-v2 — the previous release is still serving.
- Certificate pendingCheck domainshop.example.com — DNS still points to the old server.
When something breaks, ask the infrastructure.
Doctor runs deterministic checks against the node and app, gathers evidence and explains what needs attention. Repairs stay governed: only typed actions can run, with approval and audit history.
- Containerchecking…
- Port
- Environment variables
- HTTPS certificate
- DNS
- Disk
- Memory
- Server
Cause found
Your app listens on port 8080. VIBR is sending traffic to port 3000, so the health check never gets an answer.
Impact
The previous release is still serving. Nobody saw an error.
Suggested fix
Set the app port to 8080 and redeploy.
Example diagnosis. Private beta today: deploy failures, app health, Docker state, Caddy/HTTPS validation, disk, memory and node evidence. Governed repair actions are limited to explicit operations such as restarting an app or reloading the proxy — no arbitrary AI shell.
We don’t replace your AI builder. We give it somewhere safe to run.
Keep building with the AI you already use. VIBR handles everything between the code it writes and a healthy app in production.
“Create an API endpoint for customer health status.”In Cursor, Claude Code, Lovable — wherever you build.
A normal push to your branch.
One click today; automatically on push is coming. Health check before traffic moves.
New endpoint running on your own server, with HTTPS.
AI Bridge Coming
Connect the AI provider you want to VIBR with your own keys — no lock-in to one model or vendor.
Local AI, today
Want a model that never leaves your infrastructure? It’s your server — with root access you can run Ollama next to your apps right now. Your prompts and data stay on hardware you control.
AI writes code fast.
Not always safe code.
The speed that gets an app live can also ship a key in code, a query anyone can bend or a route that exposes another customer’s data. VIBR Security checks the work where deployment and infrastructure context already exist.
See an example scan — click a finding
- 1
import Stripe from "stripe" - 2
const stripe = new Stripe("sk_live_51Nx…9Qx") - 3
- 4
export async function GET(req, { params }) { - 5
const order = await db.query( - 6
`SELECT * FROM orders WHERE id = ${params.id}` - 7
) - 8
return Response.json(order) - 9
}
"some-package": "1.2.0"Why it matters
Anyone who can see this repository — a teammate, a contractor, a leaked copy — can take payments and issue refunds on your account.
The fix
Illustrative code, real VIBR Security finding types. Fixes are prepared for review rather than silently pushed to your main branch.
App security
What you ship
- Code & dependency scans
- Secrets in code
- Static and live-app checks
- Fixes as pull requests — never pushed to main
- Deploy Guard for critical findings
Runtime security
What’s running
- Live workload behaviourComing
- Container alertsComing
- Known vulnerabilities in imagesComing
- Suspicious processesComing
Access security
Who can touch it
- Roles per organization
- SSH keys per server
- Write-only secrets
- Audit log
- Approval gates & key rotationComing
Pick the protection that fits.
Security is its own fixed monthly subscription, so you can add, change or cancel it without changing the compute server underneath.
Four kinds of builders. One way to run what they build.
Whether you run a store, an agency, an AI app or a lean SaaS — the problem after the build is the same.
AI app launch
Your prototype works. Now make it production.
See how it works For ecommerce founders running AI and automationsEcommerce automation node
One private home for the automations behind your store.
See how it works For agencies and freelancersAgency infrastructure
One repeatable system for every client.
See how it works For lean SaaS founders and online operatorsLean SaaS
Own your infrastructure without becoming an infrastructure team.
See how it worksThey solve part of the stack.
We connect the stack.
Good tools exist for every layer. Here’s how they differ architecturally — and where VIBR sits.
Raw VPS
- Strength
- Cheap infrastructure and full root access.
- Trade-off
- You still operate everything yourself.
- Where VIBR fits
- Private servers too — with deploys, HTTPS, monitoring and security run for you.
Control panels
- Strength
- Great deployment and container tooling.
- Trade-off
- Built primarily for developers who self-operate.
- Where VIBR fits
- Business-friendly control, AI Doctor, integrated security and a managed option.
Platforms (PaaS)
- Strength
- Extremely easy deployment.
- Trade-off
- Platform-managed infrastructure that can become usage-based.
- Where VIBR fits
- The same easy deploy experience, on private fixed-cost compute.
App security
- Strength
- Strong application security tooling.
- Trade-off
- Another separate layer in the stack.
- Where VIBR fits
- Security built into the deploy and infrastructure workflow.
Managed app hosting
- Strength
- Much less infrastructure work.
- Trade-off
- Mostly focused on hosting ready-made applications.
- Where VIBR fits
- Built around your own AI, automations, custom apps and business infrastructure.
VIBR
Private servers, easy deploys, Doctor, security and a managed option — one control layer around the stack. Compute and software stay separately controllable.
Launch your appAlready have a server?
Keep it.
You don’t have to buy our servers to use VIBR. Connect a compatible VPS you already run and get the same control layer. Your infrastructure. Our control layer.
Pilot: we help you install the agent on your first server.
Moving shouldn’t mean rebuilding.
Already running somewhere? During early access we move it with you. Your current app stays online until the target is restored, checked and ready for the DNS switch.
VIBR is the control layer — not the owner of your business infrastructure.
No logos yet.
Everything in the open.
VIBR is early. Instead of borrowed logos and made-up numbers, here’s what is implemented in the private beta, what still needs rollout or live verification, and who you’re dealing with.
Built in the private beta
- GitHub deploys with health checks before traffic moves
- Public and private GitHub source delivery
- Domains and HTTPS for VIBR-managed deploys
- Doctor diagnostics for node and app health
- VIBR Security scans, pull-request fixes and Deploy Guard
- Private compute with root access over SSH
- Roles, write-only secrets and audit history
Next rollout
- Automatic deploy on every push
- AI Bridge with your own provider keys
- Broader governed self-healing
- Runtime security
- Customer-facing restore points
- Buying domains inside VIBR
- Broader self-service migration beyond the assisted flow
Who’s behind VIBR
- Built by MGR Products B.V., a Dutch company
- Your compute region is shown before purchase
- Business privacy and DPA terms are published before public commercial rollout — current DPA note
- Found a security issue? security@mgrproduct.nl
- Early customers talk directly to the people building it
The server price is fixed.
The rest is your choice.
Choose the compute size that fits. The server has one fixed monthly price. VIBR Control and VIBR Security are optional fixed-price software subscriptions you can activate, change or cancel independently.
- Web appNext.js, Remix, Astro · ~0.5 GB
- API or backendNode, Python, Go · ~0.4 GB
- n8n automationsworkflows & webhooks · ~1 GB
- DatabasePostgres or MySQL · ~1 GB
- Redis / queuecache & jobs · ~0.3 GB
- Scraperwith a headless browser · ~1.5 GB
- AI agentcalls OpenAI, Claude, Gemini · ~0.5 GB
- Vector databaseQdrant, pgvector · ~1 GB
- Local AI model7–8B via Ollama · ~6 GB
About 3.5 of 8 GB in use — with room to grow. Everything on one private server, one fixed price.
See the 8 GB serverRough planning estimates, rounded up — real use depends on your app. If you outgrow a server, you decide when to move up; nothing upgrades automatically.
One app or a few small automations
- 4 GB memory
- A web app, an API or a few automations
For automations and most apps
- 8 GB memory
- Apps, database and n8n side by side
Heavier workloads, teams and local AI models
- 24 GB memory
- Room for heavier workloads; local AI depends on the model and quantization
More memory, several servers, a heavier AI model or a special setup for your team or clients. Send us what you want to run and we’ll propose the infrastructure shape.
Every server includes
- Your own private virtual server
- Root access over SSH
- Backup-ready infrastructure
- Ready for VIBR Control and Security
- One fixed monthly price
- No VIBR per-request or build-minute meter
Plus, on us: the launch kit
- Personal launch reviewA VIBR engineer checks your first app — domains, secrets, database — before it goes live.
- Your first migration, with usRunning somewhere else? We move your first app with you, and your old host stays as the way back.
- A direct line to the buildersEarly-access customers talk to the people who build VIBR, not to a ticket queue.
VIBR Control (GitHub deploys, domains, HTTPS and Doctor) and VIBR Security are separate fixed-price subscriptions. You can select them while planning your stack and confirm them separately after the server payment. Without them, the private server and root access still stand on their own.
We run the infrastructure: OS, patches and firewall · Runtime, backups and monitoring · Recovery and provider escalation · not building your application.
Talk to usWe build with you: Apps, workflows and agents · Integrations and migrations · AI implementations.
Talk to usPrices in EUR per month, excluding VAT where applicable. Launch pricing: we tell you before any price changes.
Before you launch.
Do I need VIBR Control or VIBR Security?
No — both are optional. A server works on its own: you get root access and can run it yourself. Add VIBR Control if you want VIBR to deploy from GitHub, handle domains and HTTPS and explain what broke. Add VIBR Security if you want your code and access checked. You can select them while planning your server, then confirm them as separate subscriptions after the server payment — or add them later.
Do I need to know Linux?
No. You connect a repository, choose a server and press Deploy. VIBR prepares the server, builds the app, sets up HTTPS and watches it. If you do know Linux, you also get full root access over SSH — but nothing requires it.
Can I use Claude or ChatGPT?
Yes — keep building with whatever AI you like. VIBR deploys what ends up in your GitHub repository. AI Bridge (coming) will let your own AI talk to your infrastructure directly with your own keys.
Can I keep using Cursor or Lovable?
Yes. As long as your code lives in a GitHub repository (Lovable can sync to GitHub), VIBR can deploy it. You keep your editor; VIBR takes over from the moment code is pushed.
Can I bring my own VPS?
Yes. Connecting your own server is in pilot: you add it in the dashboard, we give you a one-time pairing token and help you install the VIBR Agent. The server stays yours; VIBR only adds the control layer.
Can I connect an existing GitHub repository?
Yes. Public and private GitHub repositories are supported. Private source is streamed through the authenticated VIBR control plane to your node, so your GitHub OAuth credential does not need to live on the server. Repositories with a Dockerfile work as-is; supported Node.js apps can build without one.
Does VIBR replace Vercel?
For many apps, yes: you get Git-based deploys and HTTPS, but on your own server at a fixed price instead of usage-based platform infrastructure. If you rely on edge or serverless features, a platform like Vercel may still be the better fit for that part — and that’s fine.
What happens when a deployment fails?
Your working release keeps serving. The new release starts next to it and only takes traffic after it passes its health check. If it fails, VIBR removes it, shows which step failed and why, and suggests the fix.
Can I use my own OpenAI or Anthropic API key?
Your apps can use any API key today — you store it as a secret and it’s never shown again. Connecting your own keys to VIBR itself (for AI Doctor and AI Bridge) is coming, bring-your-own-key first.
Can I run local AI models?
Yes. It’s your server: with root access you can run Ollama or LM Studio-compatible servers today. The right server size depends on the model and quantization; the larger VIBR Compute tier is the starting point we show for local-model workloads. AI Bridge will connect local models to VIBR as a first-class option.
What exactly do I pay?
One fixed monthly price per server, plus any VIBR software layer you activate. Compute, Control and Security are separate subscriptions, so you can change one without silently changing the others. No charge per visitor, request or build minute on the compute subscription, no automatic upgrades and no compute overage.
Can you manage everything for me?
Yes. Managed Infrastructure means VIBR engineers run the OS, patches, firewall, runtime, backups, monitoring and recovery for you. You keep building.
Is Managed Infrastructure the same as Managed Build?
No. Managed Infrastructure runs the infrastructure; it doesn’t build your application. Managed Build is for companies that want us to actively build apps, workflows, agents, integrations or AI implementations with them.
What happens to my applications if I leave?
Your code stays in your GitHub and builds with standard Docker — nothing is in a VIBR-only format. Before cancelling VIBR Compute you take your data with you (we’ll help). A server you brought yourself simply stays yours.
Can I migrate an existing application?
Yes. Migrations currently run with a VIBR engineer: we prepare the new server, copy the app and data, verify health, and only then switch DNS. Your current host keeps serving until then and stays available as the way back.
Now put it somewhere you control.
Deploy your app, AI or automation to private infrastructure without turning yourself into a DevOps team.
- Repository connected
- Build passed
- Health check passed
- HTTPS active