main
main copied to clipboard
Slower text generation
I've noticed much slower text generation recently with GPT 3.5 - not sure if this is a new API limitation, or related to Intellibar.
Hmm, might it be possible that you are using GPT 4? Can you take a look at your Usage and let me know. Thanks!
Hey Antonio - it's definitely 3.5, I'm also noticing more timeouts, so it may just be a local connection issue
It might be local, yes. I've experienced this as well in the past. Keep an eye on it for a while and let me know if it persists.
BTW, you are not on a free trial, right?
Hey Antonio - nope, paid version. I'm noticing that it is persisting, often with requests timing out - so I don't know if my mac is perhaps throttling data to intellibar?
Hmm, how often do these timeouts happen, every X amount of requests? I'm asking because I'm seeing very few timeouts in our error reporting. Even if all of them are from you, that's 2-3 per day.