Hi all, noob question: what is the right way to control the amount of tokens used for thinking in agent canvas? I'm currently using Kimi K3 and would like to play around with different thinking budgets. Thank you very much in advance! (BTW I'm using it via openrouter, I don't know if this changes anything)
w
wooden-umbrella-84098
07/23/2026, 8:17 PM
advanced settings in LLM Profiles has a reasoning effort dropdown 🙂
✅ 1
c
careful-queen-15960
07/23/2026, 8:31 PM
Thanks! So it's really wiring it with open router, that's nice 🙂. Thanks for confirming!
careful-queen-15960
07/24/2026, 11:01 AM
OK thanks, the difference in token usage is massive but the difference in awesomeness is negligible. 😁 (my purely subjective view)