Find answers from the community

Updated 3 months ago

Multi -modal

Hi, is there any tutorial on Multi-modal uscases using Llamaindex and weaviate? or is there any tutorial on how to store image document in weaviate using llamaindex?

10 comments

WWhiteFang_Jr

There is a tutorial for chroma in llamaindex: https://docs.llamaindex.ai/en/stable/examples/multi_modal/ChromaMultiModalDemo.html

Found one article on weaviate but not sure if llamaindex weaviate class is updated to handle this or not.

https://weaviate.io/blog/multimodal-rag

LLogan M

normally you'd just store the node in the db, and store the image somewhere else (s3? to disk?) and just have the link to the node in the metadata

PPujari12

My question is more towards if we can load and store images in the weaviate db using llamaindex class the same way as the documents?

PPujari12

its not apparently, I tried

PPujari12

is this example replicable for weaviate? - https://docs.llamaindex.ai/en/stable/examples/multi_modal/gpt4v_multi_modal_retrieval.html

LLogan M

I don't think so right now? Or at least i haven't tried

LLogan M

Give it a shot if you want

PPujari12

Its not working- getting error on index.as_query_engine() part

PPujari12

but it's working as a retriever

PPujari12

Also, can you suggest me which DB to use for Multi-modal capabilities?

Add a reply