distributev

Results 33 comments of distributev

I'm not sure if it is relevant but here is more background - I use the latest MemGPT with the Postgres storage backend. ![image](https://github.com/cpacker/MemGPT/assets/19224635/84cea925-b15a-4629-8008-6ffd54d9e509) Using existing agent Eli ``` {'append_to_text_file':...

Anybody has an idea what is happening here? I'm currently trying with gpt4 - If I use another model, maybe gpt 3.5, it might work? Why would rate limit exception...

Not sure if related but I did couple of more investigations and found this Could that be the reason? Did you see this before? ![image](https://github.com/cpacker/MemGPT/assets/19224635/035ac58a-63d7-4c60-b4c9-b0c5cd6937d6)

After I added " --debug --first --no-verify" I started to see missing pg8000 error message so I used poetry to install postgres and now I am able to chat with...

the root issue was "missing pg8000 error message so I used poetry to install postgres and now I am able to chat with memgpt" but it was incorrectly reported as...

More details - the agents which I created today do indeed use the new (less) expensive model of GPT-4-1106-preview Yesterday I was playing with an older agent which was created...

Is it possible for the displayed error to wrap/hide another error for instance "insufficient_quota" - if so, this is confusing. errors messages should be as specific as possible, not wrapped...

I start to believe that a "insufficient_quota" exception is happening but on screen it is wrapped/catched/displayed as 404 not found

It is exactly that once I added credit it started to work. The stack trace/error message should clearly display the root cause "insufficient_quota" (I saw this error message after I...

This command (taken from the above documentation page) ``` Testing out our new chatbot Let's try out our new chatbot Chaz, combined with our new user profile Bob: Shell #...