YouTube playlistImages API: 50 MB cover images from an AI image

The YouTube Data API v3 has a playlistImages resource with insert, update, list and delete methods. Insert and update accept JPEG or PNG uploads up to 52,428,800 bytes (50 MiB). The discovery document does not state required pixel dimensions, so check YouTube's help page for the cover size before you generate one.

What does the resource look like?

The PlaylistImage schema has id, kind (the fixed string youtube#playlistImages) and a snippet. The id identifies the playlist and image type. Uploads can use simple or resumable protocols.

How do I make a cover?

Generate the image at a ratio that suits a playlist tile, usually landscape, with aspect_ratio 16:9 and output_format jpeg. Because the document gives no pixel rule, confirm the recommended size in YouTube's own guidance and compare it with the dimensions Sume returns.

import json, os, urllib.request

body = {
    "model": "bytedance-seed/seedream-4.5",
    "prompt": "cover art for a cooking playlist, overhead table with fresh vegetables, bright and clean",
    "aspect_ratio": "16:9",
    "output_format": "jpeg"
}
req = urllib.request.Request(
    "https://api.sume.com/v1/images",
    data=json.dumps(body).encode(),
    headers={
        "Authorization": "Bearer " + os.environ["SUME_API_KEY"],
        "Content-Type": "application/json",
    },
    method="POST",
)
print(urllib.request.urlopen(req).read().decode())

What about AI disclosure?

A playlist image is not a video, so containsSyntheticMedia on VideoStatus does not apply here. Whether YouTube expects a label on AI-made artwork is a question for its policy pages.

What should I check?

Test the call on a playlist you own. The images guide lists Sume's side.