I’ve been getting intermittent errors on an AI Agent node citing Gemini 429 Too Many Requests. But, I can confirm it’s the only request happening on my application at that time. I’ve added wait nodes before and after, but can’t think of anything else to increase reliability… any ideas?
Hi, I might be that you are hitting this because you exceeded the number of request withing a certain time period.
Reg,
J.
2 Likes
Yes, that’s my understanding as well. But I’m only sending one… My thought is that maybe it’s too long? But even setting a token length doesn’t seem to affect it too much…
I was thinking I could throttle it somehow, but can’t come up with any good ways to do that.
Hi, Not sure if its similar to openai, but surely there must be somewhere where you can get detailed usage stats. Surely it must be in case of billing conflicrs. Which Gemini model are you using ? 2.0 flash or 2.5 pro
1 Like
I was doing 2.0 flash. I think you’re right… I was only running 1-4 per day, but each one was lengthy (a few thousand tokens). I’m going to try stringing them together and see if that helps me get my tokens down.
1 Like