How to minimize youtube-data-api v3 query quota useage?

I have a code to get video urls given a channel ID that I get from the API as well, but the main part uses the requests component.

I tried running it on a channel with 3500 videos, but for some reason running it once or twice used up 6100 credits. How is this being calculated? I don't have more than 161 requests total and they're all simple search requests.

Usage graph

def get_video_urls(api_key, channel_id):    base_url = 'https://www.googleapis.com/youtube/v3'    search_url = f'{base_url}/search'    video_urls = []    next_page_token = ''    while True:        response = requests.get(search_url, params={'key': api_key,'channelId': channel_id,'part': 'id','order': 'date','maxResults': 50,'pageToken': next_page_token        })        if response.status_code != 200:            print(f'Error: {response.status_code} - {response.text}')            break        data = response.json()        for item in data.get('items', []):            if item['id']['kind'] == 'youtube#video':                video_id = item['id']['videoId']                video_urls.append(f'https://www.youtube.com/watch?v={video_id}')        next_page_token = data.get('nextPageToken', None)        if not next_page_token:            break    return video_urls

Requests History

I tried minimizing search parameters or changing maxResults, but they don't seem to have much of an effect. Moreover the credit usage changes a lot, yesterday I capped after running it on a smaller channel twice.