Sapessii

Results 14 comments of Sapessii

Adding support for encoder-decoder models, like T5 or BART

hi @afeldman-nm! is this pr going to support also https://huggingface.co/facebook/bart-large-mnli? thank you

Handle 400 error from supabase.co/auth/v1/callback with a redirect to an error page

same here, any update or fix? thanks

bug: default settings value always overides postgres configuration when using server

had the same problem, setting these env variable worked for me MEMGPT_PG_DB MEMGPT_PG_USER MEMGPT_PG_PASSWORD MEMGPT_PG_URL MEMGPT_PG_HOST

Issue: Update VLLM to Version .5.0++, and a few suggestions

+1!

Having user with thousands of messages blocks the entire server after 2/3 messages and slow down agent reply

Hi @mattzh72, unfortunately we don’t have a clear answer on that. It looks like the issue isn’t tied to the number of messages but rather to a high number of...

Getting error when calling archival_memory_insert

Because it's calling openai api to get embeddings for the text to save

Getting error when calling archival_memory_insert

you can setup your own embeddings model locally and select it during the configuration

'invalid_grant' error upon login with Safari (WebKit)

has anyone any update on this?

'invalid_grant' error upon login with Safari (WebKit)

I have been able to fix it, on my side the problem was the service worker cache. If you are using serwist for the pwa service worker just put this...

Mistral API broken

hi! I was using the openai preset