Skip to main contentJob Priority
Sutro’s Batch API allows you to specify a priority for each job you create. For now, only two priorities are supported: 0 and 1.
Priority 0 (Prototyping)
Priority 0 is the default priority, and is primarily meant for prototyping jobs. You can expect them to complete in several minutes. By default, these jobs are soft-limited to 1000 inputs and 1 million input tokens. Please reach out to us at team@sutro.sh if you need to increase this limit.
Priority 1 (Production)
Priority 1 is currently reserved for production jobs. You can generally expect these jobs to take longer; completion times are dependent on the number of inputs, the model being used, and other factors. Generally speaking you can expect these jobs to take roughly one hour to complete.
In the future, we’ll be expanding the number of priority levels available to better serve different cost and speed requirements. 
