Data extraction from Image

[harichandana.vegesna]

Hi,

I am trying to extract data from Image using chat assistants in chat playground. Same thing I am trying to do it using nodejs code by passing the image to vector_storage and trying to extract data, its throwing error like .png file is not supported.

Error:
{
“error”: “400 Files with extensions [.png] are not supported for retrieval. See https://platform.openai.com/docs/assistants/tools/file-search/supported-files”
}

How to resolve this.

Thanks, in advance!!

[vb]

Hi!
The solution is to choose a model that has vision capabilities, like 4o, and have it extract the image features you are looking for.
Vector stores are used for different types of text formats. That’s why the error message is indicating that the file type is not supported.

Here is the corresponding part of the documentation. Make sure to switch between responses and assistants if needed.

https://platform.openai.com/docs/guides/images?api-mode=responses

[anon1374209]

you gotta read the documentation… tip: base64 is your friend

[harichandana.vegesna]

I am able to do this using chat playground assistants. But the same thing I am trying to do using nodejs code its not working. And I am using gpt-4o version