Backfill the whole table by Monday, at a price set before it starts
Classify a backlog, enrich a column from free text, backfill a table after the prompt changes. Send it all in one go with a deadline, and Otium runs it on spare computers at a price set for work that can wait.
Where AI work stalls in a data stack
-
The estimate kills the backfill
Per-request prices across millions of rows make the project look unaffordable before it starts.
-
Rate limits run the schedule
The orchestrator spends the night retrying against a limit, and the task times out anyway.
-
Nobody knows when it will finish
Downstream jobs cannot be scheduled around work with no finish time.
What Otium gives the pipeline
Hand over the backlog in one go
Submit the whole job at once; it waits safely in Otium’s queue, and the scheduler paces it, not your code.
Usage limits →A finish time downstream can rely on
Each batch carries a deadline, and the scheduler plans the hours ahead around it.
Deadlines explained →A cost before it runs
The price depends on the size, the deadline and the tokens, so a job can be costed before it is sent.
Estimating a job’s cost →Multi-step, with checks
Classify, then extract, then summarise, as one workflow with each step’s output checked before the next uses it.
How workflows run →
What your work would cost
That is $1,461 a year against gpt-6-luna’s regular price.
Assumes 2 input tokens per 1 output token, as in the costed job above. Repeated input is priced as each side bills it: free here, a tenth of the input price there, with their charge for writing the cache left out.
Questions people ask first
- What formats go in and out?
- Batch files in the OpenAI JSONL format: one request per line in, one result per line out.
- How big can a job be?
- Large. Each account has usage limits that grow with use, and they are published.
- What happens to the data?
- Nothing is used for training, each account’s data is locked with its own key, and inputs and results are deleted 30 days after the job finishes.
Cost your next backfill on Otium.
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.