Weekly TikTok trend sweep: 3 regions x 4 keywords is $1.20

A weekly sweep of four keywords across the US, GB and KR is 12 trending-videos calls and costs $1.20 per week at $0.10 each. The search takes an optional two-letter region and a window such as this-week, so each (keyword, region) pair is one call (Trending videos).

The loop

It prints how many calls it made so you can compare with the bill. query is required in production.

import os, itertools, requests

H = {"Authorization": "Bearer " + os.environ["SUME_API_KEY"]}
keywords = ["water bottle", "standing desk", "protein powder", "tote bag"]
regions = ["US", "GB", "KR"]
calls = 0
for q, region in itertools.product(keywords, regions):
    r = requests.post(
        "https://api.sume.com/v1/trending-videos/search",
        headers=H,
        json={"platform": "tiktok", "query": q, "window": "this-week", "region": region, "limit": 20},
        timeout=60,
    )
    r.raise_for_status()
    calls += 1
print("calls:", calls)

Cost over time

Notes

Without a region, a browse sends requests to the US, GB and KR trending feeds, but that no-query browse is a rebuilt-feed behavior documented as development-only until a production flag is set. For a production sweep, pass query and region explicitly as above.

The cache can serve repeat browse calls without hitting the upstream source again; do not rely on that to lower your bill, and confirm the live price in the catalog.

Fields worth setting

In production limit runs from 1 to 50 and defaults to 10; the script asks for 20. window accepts yesterday, this-week, this-month, last-3-months, last-6-months and all-time. summary_mode: metadata is included in the same $0.10 per call, and transcript currently returns metadata plus an unsupported warning.

The response is ranked public TikTok metadata: watch URLs, cover thumbnails, author handles and metrics. It does not include downloadable video files, so a sweep feeds research, not a face-swap or captions input.