Cloudron Forum

@Divemasterza This also not supported on Cloudron, I have asked before to cloudron team to have the packaged installed with mcp proxy, but still not ready yet, the solution is to use another mcp server hosted in other server like https://github.com/metatool-ai/metamcp

@Divemasterza hopefully cloduron team can make this is standalone app

May I know what is the impact of this issue?

@enerch Cloudron librechat package is not bundled with RAG API yet, so maybe we should wait next update to have it works

Latest update already works well, the thing that we need to add to config.php this configuration " 'allow_local_remote_servers' => true," . This line of configuration is especially to allow connection to LLM or Inference engine like openAI and other openAI API compatible

looking forward to latest update

@Valexico Sure, to make it more efficient,but somehow RAG knowledgebase deem necessary for business usage as RAG can enhance model to understand the business context of the company using tool like librechat, without RAG knowledgebase, the model will only have vast general knowledge

But without RAG API, somehow we can nor "chat" with files, that's from understanding about how important to have RAG API packaged with core Librechat. Pleaset let us know if we can upload files to Librechat without RAG API, maybe I miss something

@james @girish I think untill the RAG API not bundled with initial package in Cloudron, the librechat couldn't function well as Generative AI, I hope Cloudron team can bundle the package in next update

@luckow The RAG is quite different with Meili Search, eventhough they share similar purpose but the RAG API is intended to enable librechat to be able to upload files and processed by embedding model

@nebulon the documentation for RAG API can be found in https://github.com/danny-avila/rag_api

As seen that the pgvector can't be used if the RAG API not installed and running. This RAG API is somehow critical part, if it's not running then the librechat can't have capability that openwebui has for files upload, processing files to embedding model and so on

Thanks for the info, but I couldnt find RAG API that can connect to pgvector, am I missing something or it's not packaged yet? The documentation in Cloudron still not mentioning this

It seems the package is not shipped with RAG API ?

I wonder if the existing Librechat stack in Cloudron already include with PostgreSQL pgvector?

@apesorguk you don't need to revert to old backup, with the latest update just enter the information of your openwebui instance. You enter with www.yourdomain.com:3000 as admin then go to setting > General and then enter your domain

To solve this issue please take a look here as the solution. It might not be the best solution that you expect, but so far, this can resolve your issue

I once post in the forum that I hope in the next version there will be feature to customize reset password link (and text if possible) in case the cloudron instance user directory linked to another cloudron or other user directory providers. The latest version of Cloudron lack of this feature which I think is important to make it easy for regular user who dont remember the user directory url (which mostly dont remember)

Vote for this two times

I have plugins regarding to this feature off for a while in Nextcloud, till the issue is found and hopefully resolved

@james Have you reproduced the issue? DId it happen in your end?