Can I submit a job today?

Not yet. E31 Network is a technical preview, and nothing has shipped. There is no submission path, no queue accepting work, and no meter running for anyone. What the site publishes is the design of the thing, written down before the thing exists, so that teams with the kind of jobs it is meant for can read it and tell us where it is wrong.

That is a deliberate choice rather than a delay. A compute marketplace built around batch jobs has to get the shape of a job right before it takes the first one, and the shape is easier to argue about on a page than inside a half-built system.

What is published so far

The unit of the network is the job, and a job declares four things. The work: what runs, whether a training run, a batch of inference, a processing pass over footage, or a set of simulations. The data: what the job reads, staged for the job. The request: how much compute, for how long or until done, and when it must be finished. The output: what comes back, whether a model, a set of results, processed files, or a report.

A job would then move through five steps. Submit, where the four parts are described. Rent, where compute is held for the job and the meter starts. Run, where the job runs unattended and isolated from other jobs. Output, where the results are returned to the submitter. Release, where the compute goes back to the pool and the meter stops. Payment covers rent and run only. The full walk-through is on how it works.

What is not published

Rates. None exist today, and the site says so wherever the subject comes up. The question of cost has its own article at what does it cost.

The submission mechanism. The design describes what a job declares, not the tool a team would use to declare it. Data retention and deletion specifics are also not yet published, and neither is where the capacity comes from. The program keeps a list of what it has not solved, and that list is meant to be read, not hidden.

What a team can do now

The most useful preparation is to write down a real job in the four parts above. Pick the retraining run you keep postponing, or the backlog of footage sitting on a drive, and describe its work, data, request, and output in a paragraph each. If the request part is hard to write, that is informative: it usually means the job's deadline or its appetite for compute has never been pinned down.

The examples page shows four illustrative requests written that way. The design partners page describes the kind of team the program would want first, and states that intake is not yet open. When that changes, the page would change with it. Until then, the honest answer to the title is the short one.