Pipeline Builder Use LLM Node Limits

Are there any upper bound constraints on the number of rows that can be cached when using the Skip Recomputing Rows feature of the UseLLM Node?
On a related note, are there any upper limits to how many rows an LLM Node can process (separate from Language Modeling Service Limits)?

There is not a hard upper bound constraint for the Skip Recomputing Rows feature! If you are ever having scale issues feel free to flag to us.

There may be rate limits coming from our Language Model Service but there’s also no hard ceiling here

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.