Pricing and Workspace Sizing

Open Vassal offers hosted agent plans with different runtime capacity levels. The right plan depends on the kind of work the agent will do and how much headroom you want.

Start with the smallest plan that fits the workload, then move up if the agent needs more compute or memory.

Workspace add-ons let you increase the persistent storage available to the agent. This matters when the agent needs to keep a larger working set of files or outputs over time.

The goal is to size storage based on the actual workspace footprint, not just to maximize it by default.

Choose capacity based on the job, the expected file volume, and whether the agent will be part of a team workflow or a solo workflow.

You can begin with a straightforward configuration and expand only when the workload proves it is necessary.

In practice, the cleanest signal is operational friction. If the agent is regularly constrained by memory, workspace size, or runtime headroom, then it is time to move up.

If the workflow is still undefined, adding more capacity usually does not fix the actual problem. Better task boundaries and better files do.

Start by estimating how much of the agent's work is compute-heavy versus context-heavy. Compute-heavy work tends to push you toward a larger runtime. Context-heavy work tends to push you toward more workspace.

Then decide how costly slowdowns would be. If the agent supports something business-critical, extra headroom is often cheaper than repeated operational friction.