Possible for a user to upload and anlyze images? (jpeg,png etc)

hi! i wonder if its possible for a user to upload and anlyze images? (jpeg,png etc)

3 Likes

Currently, an end-user (someone using your tool) is not able to upload a picture. Only word docs, PDFs, CSVs, and similar documents. You can expect this to change in the next 2-4 months though.

3 Likes

Would be interested in this as well.
I’ve used this with Chatgpt before. One idea is to have a user submit outfits their wearing and the AI will judge if its appropriate for an occasion.

2 Likes

Is there any update on this? Would be great to upload jpgs etc and have ai rewrite or understand whats involved in the image

2 Likes

Hi Dan, this feature is coming very soon. We actually quietly implemented it this past weekend and it’s live in the builder. So actually it is ‘secretly’ available to users as long as you’re using GPT-4o model. If you have “end-user upload” enabled you can upload an image through the paperclip icon. We are debugging it this week before rolling it out officially. If you want to give it a whirl, go ahead!

When the user downloads an attachment, it remains in the conversation forever. How can I disregard the p-j if the user starts a new topic/uploads another attachment?

Are you referring to when the user uploads an attachment into a chatbot?

Yes exactly. Attachement in a chatbot

When an end-user uploads a document or image into the chat, it is only used for that unique conversation session. Other uses will not see that document, the document is not stored, and if the same end-user refreshes the page with a new conversation they will not see the document.

Within the context of a conversation, an end-user can upload a new document at anytime too.