A model on your own machine
We install an open-weight language model on a computer or server you own, sized to the hardware you have, with a simple way for your team to use it.
For private AI projects
Run AI models on hardware you control.
We set up open-weight language models on machines you own and connect them to your files and tools, so the data stays with you. Every project gets a written quote first.
- Written quote first
- You own the code
- Replies come by email
Direct with the developerNo account manager in between. You talk to the person writing the code.
You own itThe code, plugin and site we deliver belong to you.
Based in AustinCall (830) 534-4148 or email us.
What we build
Here's what we set up for private AI.
A good first project is one job you'd like a model to help with, plus a clear list of what it's allowed to see.
Search across your own files
Ask questions and get answers that link back to the document they came from, without uploading those documents to an outside AI service.
A connection to your tools
The model can read from your help desk, database or shared drive, with read-only access unless you decide otherwise.
More examples
We can also build these.
Each one starts with a list of the data the model is allowed to see.
- A private chat assistant for staff, running on a machine in your office
- Summaries of long documents that stay on your network
- Sorting incoming support tickets into queues for a person to check
- Sending simple questions to a small local model and harder ones to a larger model you choose
- A test set that checks the model's answers against your own before anyone relies on them
Prices
Start with one job on one machine.
There's no set starting price for this work, because it depends on your hardware and your data. Describe the job and we'll price it before anything starts.
Written quote
Local model setup
A model installed and running on hardware you own, with a way for your team to use it.
Written quote
Document search
Questions answered from your own files, with a link back to each source.
From $2,000
WordPress connection
A custom plugin that lets your WordPress site or its admin use the model you run.
Our work
Our apps keep typed text and files on your device.
Each of these does its work in your browser, and what you type or open isn't sent to Alcor. If you talk instead of typing, your browser turns speech into text, and Chrome or Edge may send that audio to Google or Microsoft to do it.
AlcorTranslate
Translates a conversation both ways with the browser's built-in, on-device translation, and needs no account or API key. It currently runs in desktop Chrome 138 or later.
Open AlcorTranslateAlcorPhoto
Edits images in the browser. Your files stay on your device, and there's no account.
Open AlcorPhotoAlcorVideo
Edits and renders video in the browser, so your footage stays on your device.
Open AlcorVideoHow it works
Here's what happens after you write to us.
- 01
Tell us what you need
Use the form below or email alcor@alcor-industries.com, and say what it should do and by when.
- 02
Get a written quote
Once we understand the job, we send the price and what's included in writing. No custom work starts before you have it.
- 03
We build it
We build what the quote describes and let you know when it's ready.
Questions
Common questions
What hardware do we need?
It depends on the model and how many people use it. Tell us what you have now, and we'll say whether it's enough or what to add.
Can we use a cloud model for some tasks?
Yes, if you want to. The quote will say which tasks go to which model and what data each one sees.
Will it act on its own?
Only as far as you allow. Anything that sends, deletes or spends money can wait for a person to approve it.
AI infrastructure
Tell us about the job.
A sentence or two is enough. Tell us what it should do and add a rough budget if you have one.
You can also email alcor@alcor-industries.com or call (830) 534-4148.
Pick another page