AI limits or throttling

DEVONthink Artificial Intelligence
1

Great work on the AI API in DT4, thank you for those very useful features.

I use Anthropic Sonnet 4.5 with a 5x Max subscription and Anthropic limits to 50 query per minute.

If I run a batch process that needs multiple queries per document and the number of document is too high, some processes silently fail.

Is there a way to set a kind of limit or throttling in DT4 so it doesn’t send too much request per minute ?

Thanks for your help

1 Like
2

Welcome @jnury and thanks for the kind words.
No there is no throttling controls you can access.

What specifically are you doing with your batch process?

3

DEVONthink honors the according responses by AI providers and throttles and retries requests automatically but only few times.

Was anything of interest logged to the Window > Log panel or to the file ~/Library/Application Support/DEVONthink/Chat.log?