I’m reaching out about an issue with the Telegram Download attachment node, which downloads image attachments in low quality. This prevents OCR software from accurately processing text, impacting workflows that rely on clear images.
Could you look into this and suggest any possible fixes?
Thanks for your help!
Hi! The issue you’re encountering is caused by the fact that the Telegram API returns several preview sizes for each image, and by default, the smallest one is selected.
When an image is uploaded to Telegram, it typically generates multiple versions of that image in different sizes - most often three. This is why you’re seeing an array with several entries for the same photo.
As you can see, if you check the path, you’re receiving three sizes of the same image:
And when you select the variable, the path defaults to the [0] element — which is the first and smallest one:
Just change the path to access the third element instead (indexing starts from 0, so you write 2):
In that case, everything will load in its original resolution:
This topic was automatically closed 4 days after the last reply. New replies are no longer allowed.
