Vertex AI Agent Platform
I’m having trouble finding rate limits for models. Ive looked in the usage center and on my side I basically see either non existent limits of like 5 per minute or unlimited per minute.
Anyway, I’m hitting 429 every so often when sending lots of quick requests to 3.5 flash lite, im talking like maybe 100 to 200 in a minute.
Does this sound like an acceleration problem? There’s no way rate limits are that low are they? We are tier 3 on ai Studio, does that translate over to the enterprise side?
I liked running these requests on vertex just because it’s so much faster than ai studio…but I’m going to need to scale this up to like 500 a minute for 15 minutes straight soon…