Feed mantid the data and files you already have. Train an ML model that predicts and forecasts, or train a custom LLM for general AI.
Free. No account required. Your data never leaves your computer.
Drop in your data. mantid trains the ML model. #
Bring what you have.
A spreadsheet, an export, a folder of PDFs. mantid reads it and says what it can do with it.
Pick a question.
mantid suggests questions your data can answer (or type your own).
Go and make a cup of tea.
It trains on your machine while you relax. mantids are patient.
Get a verdict.
mantid reports back on the custom AI model.
Put it to work.
Drop in next week's data to get answers, or export the model into your own project.
| weekday | days_ahead | past_no_shows | sms_sent | no_show |
|---|---|---|---|---|
| Mon | 21 | 2 | no | yes |
| Tue | 3 | 0 | yes | no |
| Tue | 14 | 1 | yes | no |
| Wed | 30 | 3 | no | yes |
| Thu | 1 | 0 | yes | no |
What do you want to know?
Bad models get eaten. #
Every model is measured against the simplest guess anyone could make from the same data. mantid tells you whether it beat that guess, and what would help if it didn't.
Retrain your models with versioning.
Nothing is overwritten. Go back to any version you liked better.
Every verdict comes with a fix.
Add data, correct a mistake, set a rule, then run it again.
This model doesn't beat the simple guess.
The simple guess says nobody leaves. It is right 97 times in 100. Your model is right 96.
What would help: only 38 of your 1,200 customers left, so add a column like last_login and bring two years of customers instead of one.
Bring fuzzy data. Let mantid figure it out. #
mantid interprets your data for you.
Gyms
Which gym members are about to ghost us?
cancelled Dental practices
Who's going to skip their check-up?
no_show Bakeries
How many croissants do we bake on Saturday?
units_sold Hotels
Which bookings will cancel last minute?
cancelled Workshops
Which machine breaks down next?
failed_30d Factories
Which batches fail inspection?
failed_qa Software teams
Which trial users will actually pay?
converted Sales teams
Which leads are worth a phone call?
won Restaurants
How many covers on Friday night?
covers Charities
Which donors will give again this year?
gave_again Support teams
Which support emails are actually urgent?
urgent
Your oddly specific question goes here
If you've got the data, mantid will tell you if it can be answered.
Train it on your documents. Then ask it anything. #
Drop in your documents and data. mantid trains your own LLM on them, then you use it like any assistant, on your own machine.
Give it anything
PDFs, Word files, emails, notes and spreadsheets. As much as you have.
Ask your documents
Every answer from your files shows the file it came from, and "not from your documents" beats a guess.
Put it to work
Tell it what to do, hand it a new file to read, and get text or a table back.
Tested before you use it
mantid checks it against the untrained model on questions from your own files it never saw.
Can staff carry over unused holiday?
Many companies let staff carry over some holiday. Check your employee handbook or ask HR.
Yes, up to five days into the next year, if your manager agrees in writing before 31 December.
Holiday policy.pdf, p.3
Pricing. Nom nom. #
Free
You read that right.
Train as much as you like on your own machine. Nothing to pay and no account.
- Train as many ML models and LLMs as you like on your data.
- Use them on new data, files and folders of documents.
- Export your models for DIY deployment.
- Call your models in your AI tools with a local MCP.
- Your data never leaves your machine.
A Mac with Apple Silicon (M1 or later), macOS 14 or later. ML models need 8 GB of memory, LLMs need 16 GB. Anonymous usage data is on by default and can be switched off.
Hosted with API & MCP
Use your models in any platform, AI tool or vibe coded app. High uptime, low cost, usage based fees.