Hello,
I created a scenario where I upload a Google Docs file using the OpenAI “Upload File” module. After that, I want to send the content to an assistant using the “Message Assistant” module.
The module doesn’t display an error message. However, the output I receive is: “It seems like we are unable to read your file.” If tied to add the file ID somehwere in the module I got htis error message : “[404] Vector store with id ‘file-LwT9NAYfnrrgjXALKHKS6D’ not found.”
What could be causing this issue, and how can I fix it? Where can I add the file data into the “Message assitant” module ?
Thanks in advance!
EDIT: I thought you were trying to directly upload the google doc file straight into the assistant. I’ve now noticed you are uploading to OpenAI first, so the below will not help.
I had a similar issue with accessing Google Drive files.
Try using this when calling the file: https://lh3.googleusercontent.com/d/[Insert File ID]
You need a specific google drive path.
Hey !
Thanks for your help but your solution doesn’t seem to work.
best,
I’ve just tried building the same thing. The file that I am uploading is showing as available for Code Interpreter, but not file Search Resources. If I map the File-ID to the Interpreter and not the File Search, it seems to work.
Hey,
I tried your proposition and I don’t have an error message but It says that OpenAI couldn’t access the file (sorry it is in French).
Do you know how I can fix this?
Thanks for your help!
I am having the exact same problem. It is weird that it works with code intepreter and not file resouce. I suspect that this is a bug in the implementation of the Module. Is it possible to get someone from MAKE to take a look at this?
Hey! Can the OpenAI module access the file? Because in my case, it doesn’t seem to be working.
The File Search Resources is after a Vector Store and not an individual file, this is why its not working. Here is the explanation from OpenAI:
" Vector stores. Vector Store objects give the File Search tool the ability to search your files . Adding a file to a vector_store automatically parses, chunks, embeds and stores the file in a vector database that’s capable of both keyword and semantic search. Each vector_store can hold up to 10,000 files."
I’ve also noticed there appears to be a module to Add Files to Vector Store. I’m thinking we try this and then use the Vector Store in the Assistant Module. When I get a chance I’ll try it out.
1 Like
Do you know how to set up that module? I’m kind of stuck now ![:frowning: :frowning:](https://emoji.discourse-cdn.com/twitter/frowning.png?v=12)
Thanks a lot for your help.
Best,