Hey everyone, I've been putting the cloud version of BabyAGI through its paces for a few weeks now, mostly loving the automated task handling for my side projects.
But this week, I've started hitting a frustrating number of API timeouts. It seems to happen most often on longer, more complex task chains—like when I ask it to research a topic and then draft a summary. The process will just hang and eventually fail with a generic timeout error.
A couple of patterns I've noticed:
* It's worse during what I assume are peak hours (US afternoons/evenings).
* Simpler, 3-4 step tasks usually complete fine.
* I'm not hitting any obvious rate limits on my end.
I'm using the default settings with GPT-4 as the LLM. My workflow is basically feeding it an objective and letting it create its own task list.
Has anyone else run into this? I'm wondering if:
* This is a known scaling issue with the cloud service.
* There are specific task chain structures that are more prone to timing out.
* Anyone has found a workaround—maybe breaking objectives into smaller chunks manually?
Kinda kills the "set it and forget it" vibe if I have to babysit it 😅. Would love to hear if others are seeing the same thing or if it's just my setup.