Run the whole corpus on a model that will still exist when the reviewer asks
Labelling, coding, screening or extracting at the scale of a study, on a grant budget. Otium runs open models anyone can download, at prices set for work that can wait, so the run fits the budget and the methods section names a model a reader can get.
The usual problems
-
The model in your methods section is gone
Hosted models are retired and changed on their vendor’s schedule, and a result you cannot run again is hard to defend.
-
The budget is fixed; the bill is not
Prices set per request for a person waiting turn a corpus into a guess, and a grant does not stretch.
-
Where did the text go?
Ethics boards and data agreements ask who sees the documents and whether they train anything.
How Otium fits a study
Open models anyone can get
Each size runs a published open-weight model whose weights are public, so the model you cite can be run again by anyone. When a size moves to a newer model, we tell you first.
Open and closed models →A price you can put in a budget
The deadline you choose sets the rate, and the playground prices a pipeline before you send it.
Price a pipeline →Never trained on, deleted on schedule
No AI company sees the text: we run the models ourselves. Nothing trains on it, and it is deleted 30 days after the job finishes.
Privacy and security →Files in, results out
Send a batch file in the OpenAI format your scripts and notebooks already write, and get one result per line back.
Structuring a batch job →
Questions people ask first
- Can my institution pay by invoice?
- Accounts are prepaid by card today. If your institution needs an invoice or a purchase order, write to [email protected].
- Which models do you run?
- Each size names its model on the pricing page; all are open-weight models published by their makers.
- How long does a run take?
- You set the deadline: within a day, or within a week for less. A corpus is sent in one go and comes back whole.
- Can I try it first?
- Yes: an account and the demo model are free, so you can check the output on a sample of your corpus before you spend anything.
Try it on a sample of your corpus.
Otium is in beta. Leave your email and we’ll send you a link to create your account. Then point your OpenAI-compatible client at our base URL and send your first batch.
Free to try on our demo model. No credit card, no spam: we email you a link to finish.
Otium is in beta and open to everyone. Already have an account? Sign in.