Hey there @balloob, mind taking a look at this issue as it has been labeled with an integration (openai_conversation) you are listed as a code owner for? Thanks!

_{^{(message by CodeOwnersMention)}}

openai_conversation documentation
openai_conversation source
_{^{(message by IssueLinks)}}

I have the same problem. When i remove the '-' before the list it does upload but it says the following:

Error generating content: Error code: 400 - {'error': {'message': "Invalid 'input[0].content[1].file_id': '/config/images/voordeur.png'. Expected an ID that contains letters, numbers, underscores, or dashes, but this value contained additional characters.", 'type': 'invalid_request_error', 'param': 'input[0].content[1].file_id', 'code': 'invalid_value'}}

If i understand correctly, openai expects you to upload the file first and then use the fileid that it returns when uploaded?

I am running into the same thing. Also the documentation needs to be updated. It is no longer image_filename:

+1 please update the documentation

Thanks for a great integration.

I am having the same issue with filename error code: 400. I tried different variations on the file name without success.

Upon further research this might be caused by the OpenAI API requiring a filename rather than a path. The use of a path with the "/" symbols may be what is causing an error.

Seeing the same error as Melantrix above.

action: openai_conversation.generate_content
data:
  filenames: /config/snapshots/snapshot_camera.jpg
  config_entry: blah
  prompt: |2-
        Very briefly describe what you see in this image from my doorbell camera.
        Your message needs to be short to fit in a phone notification. Don't
        describe stationary objects or buildings.

Just popping in to say I'm having this exact issues as well.

action: openai_conversation.generate_content
data:
config_entry: 01JS0P7YARPTKGD9KXKQEA7D3A
prompt: You see trash can?
filenames: /config/www/snapshots/trashsnapshot.jpg

Error:
Failed to perform the action openai_conversation.generate_content. Error generating content: Error code: 400 - {'error': {'message': "Invalid 'input[0].content[1].file_id': '/config/www/snapshots/trashsnapshot.jpg'. Expected an ID that contains letters, numbers, underscores, or dashes, but this value contained additional characters.", 'type': 'invalid_request_error', 'param': 'input[0].content[1].file_id', 'code': 'invalid_value'}}

I'm having the exact same issue.

Same here!

I'm also impacted.

Exact same issue here. Can't use images.

+1 for a fix, running into the same problem

Wow @balloob himself an integration owner. :) Hi!

I hope this can be fixed soon, as this is one of the main LLM integrations.

The problem is:
https://github.com/home-assistant/core/blob/dev/homeassistant/components/openai_conversation/__init__.py#L140

The file_id provided in service is a path and it may contain chars that are not allowed by OpenAI.

I think we can simply generate random file id or just put 1 into that field.

This field actually seems to be optional:
https://github.com/openai/openai-python/blob/main/src/openai/types/responses/response_input_image_param.py#L21