Scaling AI
$1/mo base
- Plus about $0.02 per CPU-minute used
- Scale to zero when idle
- Serverless AI inference
- Best for development and bursty workloads
A private platform provider.
Serverless AI hosting that scales to zero.
Run Python apps, REST APIs and AI models on managed containers. Start serverless from $1 a month and pay by the CPU-minute, or take a dedicated container with fixed CPU and RAM.
You choose the AI. WrkMate runs the server for you in plain English.
- Scales to zero when idle
- Real MySQL database
- One bill
Compute tiers
- Scaling AI
- $1/mo base
- 0.5 CPU + 2GB
- $25/mo
- 1 CPU + 8GB
- $40/mo
- GPU compute
- Talk to us
First month free. WrkMate, Apache and DirectAdmin on every tier.
How it scales
You pay for the minutes it works.
Scaling AI starts at a $1 a month base. On top of that you pay about $0.02 for each CPU-minute your app uses. When nothing is running, nothing is metered.
For a sense of scale: 100 CPU-minutes of work adds about $2 to the $1 base. That is arithmetic on the published rate, not a quote for your workload.
Scaling AI
- Base
- $1/mo
- Usage
- about $0.02 per CPU-minute
- When idle
- scales to zero
- Includes
- Apache, DirectAdmin, WrkMate
Compute tiers
Pick your resources.
Apache, DirectAdmin and WrkMate come on every tier. First month is free.
0.5 CPU + 2GB RAM
$25/mo
- 0.5 vCPU dedicated
- 2GB RAM
- 1GB NVMe storage
- Always on and predictable
- Best for small APIs and inference endpoints
1 CPU + 8GB RAM
$40/mo
- 1 vCPU dedicated
- 8GB RAM
- 1GB NVMe storage
- Handles heavier workloads
- Best for production apps
GPU compute
Talk to us
- GPU capacity for model inference
- Sized to your model and traffic
- WrkMate included
- We spec it with you
What runs here
From an idea to something you own.
AI inference endpoints
Put your model behind an endpoint that scales to zero when idle and bills for the minutes it works.
React and Node.js apps
A front end and an API on managed containers, wired to a real MySQL database you control.
Python services and APIs
REST APIs, background workers and scheduled jobs, with Apache and DirectAdmin handling the plumbing.
WordPress next to a custom app
Run a WooCommerce store and a separate app on one account, with one bill and one assistant.
Deploying
Three steps from a container to a running app.
The same steps the AI Platform page describes, in order.
- 01
Create the database
Ask WrkMate for a MySQL database. It returns the credentials and writes them into your config.
- 02
Point a domain at it
Add a subdomain, set DNS records, install SSL and create mailboxes, all in plain English, in the same account as the app.
- 03
Deploy and run it
Push your app, connect it to the database and let WrkMate run the server. Prefer your own AI? Bring to Local hands it the same control from your computer.
Questions
Plain answers.
What does scale to zero mean?
The container scales up when your app is working and down to nothing when it is idle. You pay for the compute minutes you use, not for a box that sits waiting.
How much does serverless AI hosting cost?
Scaling AI is $1 a month as a base, plus about $0.02 per CPU-minute used. Dedicated containers are $25 a month for 0.5 vCPU and 2GB RAM, or $40 a month for 1 vCPU and 8GB RAM. First month is free.
Serverless or a dedicated container?
Scaling AI suits development and bursty workloads. The 0.5 CPU and 2GB tier suits small APIs and inference endpoints that stay on. The 1 CPU and 8GB tier suits production apps.
Do you offer GPU hosting?
GPU capacity for model inference is sized to your model and traffic. There is no published price. Talk to us and we will spec it with you.
What can I run on it?
React, Node.js, Python, plain static sites, WordPress and WooCommerce, on managed containers with Apache and DirectAdmin underneath.
Is there lock-in?
No. You get a real MySQL database you own, a real domain and standard tooling. It is your code, so you can move it whenever you like.
Serverless AI hosting
Start at $1 a month. Scale when you need to.
Dedicated containers start at $25 a month. First month is free and WrkMate is included on every tier. Need GPU capacity? Talk to us.