Deep Gandhi

@mcharters To answer your initial questions, you can upload first documents without the need to create any directories. To use a specific document for rag and index it you should...

Hey @mcharters, Just writing back to you since we added some new things. Transformer Lab now uses markitdown as the document processing engine. You can upload a lot more and...

This seems quite weird, I think somehow the embedding models on your older embedding and this newer one got mixed up. Would it all be possible for your to delete...

Could you once try clicking the "Reindex" button? I think it didn't reindex automatically for some reason which I will check out Edit: I checked the auto indexing on RAG...

Hmm this means there is not enough space on the GPU to handle everything as it just sends your query and two of the most relevant contexts from your RAG...

Yeah its the same issue under a different error. Your model worker gets killed because of the memory issues

Hi thanks for reporting this! We will fix this asap and also post the new setup.sh so you can continue using this until we do a release

The PR is open (transformerlab/transformerlab-api#329) for a fix but here is what you need to change the setup.sh to: ``` uv pip install llama-index==0.12.38 uv pip install llama-index-llms-openai-like==0.4.0 uv pip...

Hey @bumpyetching0a, Thanks so much for providing the file, would it be possible for you to open a PR and submit this file so we can test things out?

Hi @bumpyetching0a, We have moved our torch version to 2.7 which does not support Intel Macs as shown in the [Pytorch Announcement](https://dev-discuss.pytorch.org/t/pytorch-macos-x86-builds-deprecation-starting-january-2024/1690). To maintain system uniformity, we have also decided...