Hey OpenAI Team,
With the introduction of ChatGPT-4o, we’ve unlocked some incredible capabilities:
- Input: Upload general files like spreadsheets to use as templates, partially filled files, or files with errors. So not part of the knowledge based, but active documents to work/edit with.
- Output: Download various types of documents as a result of given operations. For instance, generating a zip file from a collection of generated AI files or providing the finished template that was originally uploaded.
Currently, the Assistant v2 supports attachments as part of the RAG environment for the agent but only for images.
I’m curious if there’s an estimated timeline for when Agents will be able to handle input files that are not part of the RAG and are not limited to just images?
Thanks for all your hard work and dedication to enhancing the capabilities of AI for the community!
Best regards,