How to Increase Max Cost for Claygent Column and Rerun for Error Cases

Dean F. · 2024-01-26 15:23:40.720681+00

is there a way to increase the max cost for a claygent column and run only for those that errored due to cost?

How to Increase Max Cost for Claygent Column and Rerun for Error Cases | Clay

12 comments

· Sorted by Oldest

Clay Team
APP
·
·
Hey hey! Can you drop the URL for your table here? (if applicable) 🫶 Don't worry, only admins will be able to open your table 😉
Clay Team
APP
·
·
Hey Dean! Great question, but to ensure you can save and prevent re-runs for already enriched rows, I would duplicate this column, change the settings, and give it the conditional formula to run for those that didn't have any results.Will ask if it's possible, but currently, if you modify any of the settings or inputs, the integration will consider that an entirely new request and could re-run existing results.
Dean F.
·
·
ok issue is with duplicate column not sure how to get it to recognize the rows that "errored" due to cost
Dean F.
·
·
https://app.clay.com/workspaces/154083/tables/t_9ZxGIwTqYWD6/views/gv_xLyh8IK9Ihzg
Dean F.
·
·
vs run condition not met
Dean F.
·
·
it would be nice to have a waterfall that slowly increases the "max cost" on its own only against the errored rows for claygent
Clay Team
APP
·
·
Yep, that's a tricky one! You can still give it a conditional to only run for those that didn't return anything in the "result" field, i.e. only run if it's empty but of course, as you said, the issue is that not all errors are related.If majority of cases are then that would cover most of these for now.
Clay Team
APP
·
·
I'll submit a request so the team can check it out and let us know if it would be possible or a workaround for that.
Dean F.
·
·
ok that would be good
Dean F.
·
·
thanks
Adam E.
·
·
Like Arturo O. said, there is a roundabout way to do this which is to extract the field that indicates it returned early because of cost, duplicate the Claygent column, then add a conditional formula based on the "returned early because of cost", and run that column
:sparkles-clay:1
Adam E.
·
·
Dean F. no built in way to do this - but we wouldnt recommend slowly increasing the max cost - you'd just be wasting your own OpenAI credits It won't run differently based on the max cost, it will just stop trying to find the data earlier. For the ones that didn't hit the max cost, even if you set a higher max cost, they would've cost the same amount. Does that make sense?

Clay Team
APP
·
·
Hey hey! Can you drop the URL for your table here? (if applicable) 🫶 Don't worry, only admins will be able to open your table 😉
Clay Team
APP
·
·
Hey Dean! Great question, but to ensure you can save and prevent re-runs for already enriched rows, I would duplicate this column, change the settings, and give it the conditional formula to run for those that didn't have any results.Will ask if it's possible, but currently, if you modify any of the settings or inputs, the integration will consider that an entirely new request and could re-run existing results.
Dean F.
·
·
ok issue is with duplicate column not sure how to get it to recognize the rows that "errored" due to cost
Dean F.
·
·
https://app.clay.com/workspaces/154083/tables/t_9ZxGIwTqYWD6/views/gv_xLyh8IK9Ihzg
Dean F.
·
·
vs run condition not met
Dean F.
·
·
it would be nice to have a waterfall that slowly increases the "max cost" on its own only against the errored rows for claygent
Clay Team
APP
·
·
Yep, that's a tricky one! You can still give it a conditional to only run for those that didn't return anything in the "result" field, i.e. only run if it's empty but of course, as you said, the issue is that not all errors are related.If majority of cases are then that would cover most of these for now.
Clay Team
APP
·
·
I'll submit a request so the team can check it out and let us know if it would be possible or a workaround for that.
Dean F.
·
·
ok that would be good
Dean F.
·
·
thanks
Adam E.
·
·
Like Arturo O. said, there is a roundabout way to do this which is to extract the field that indicates it returned early because of cost, duplicate the Claygent column, then add a conditional formula based on the "returned early because of cost", and run that column
:sparkles-clay:1
Adam E.
·
·
Dean F. no built in way to do this - but we wouldnt recommend slowly increasing the max cost - you'd just be wasting your own OpenAI credits It won't run differently based on the max cost, it will just stop trying to find the data earlier. For the ones that didn't hit the max cost, even if you set a higher max cost, they would've cost the same amount. Does that make sense?