YouTube search.list: 100 calls a day, and what Sume searches

Since June 1, 2026 the YouTube Data API charges search.list to its own bucket with a default of 100 calls per day, one unit each, separate from the 10,000 units shared by other endpoints. For finding reference Shorts that is a tight budget, and Sume cannot widen it: Sume's trending search covers TikTok metadata only, not YouTube.

YouTube facts come from the quota page and the revision history, read on 2026-10-02. Sume facts come from the trending videos docs.

What changed for search.list on June 1, 2026?

The revision history says a granular quota system began that day: videos.insert and search.list are charged separately, and every other method stays in the existing bucket. The quota page now reads: a project that enables the API starts with 100 search.list calls, 100 videos.insert calls and 10,000 units per day combined for all other endpoints.

Quotas reset at midnight Pacific Time. The page we read does not explain how to request more, and it points you to the Quotas page in the Google Cloud console to see your usage, so we do not describe an extension process.

How should I spend 100 searches?

Treat each call as a scarce research query. A search that returns a result list costs one call whether you read one result or all of them, so ask for the full page and keep it.

• Cache every response on your side, keyed by the query text and the date.
• Write down the five or six queries that matter before you start, not while exploring.
• Save the video IDs you want to study and read them later through other endpoints, which draw on the larger shared bucket.
• Keep search separate from uploads: they are different buckets, so a heavy search day does not eat your 100 upload calls.

What does Sume search, and what does it not?

Trending videos returns ranked public TikTok video metadata: watch URL, cover thumbnail, author handle, metrics and optional summaries. The platform field supports only tiktok. On production the query is required and limit runs 1 to 50, default 10; the rebuilt browse feed is on for the dev environment only.

It is a paid research utility, not a generation model, and it does not download or mirror videos. There is no YouTube search in it. If you need a YouTube reference, use the YouTube API within the bucket above or the YouTube site itself.

Can Sume read a YouTube link I found?

Not directly. Sume's media tools read files that are already on media.sume.com, and off-site URLs are rejected at admit, so a YouTube page link is not a valid video source for video inspect. Our post on the 422 unsupported_source error covers the legacy analysis route. Use only footage you own or have the right to use.

What is a sensible research routine under the cap?

Plan the week in advance: one or two search calls per topic, run once, cached, with the results saved as a list of video IDs and notes. Use the larger shared bucket for follow-up reads of videos you already found, and use platform-native browsing, which spends no API quota at all, for exploration.

For TikTok-side research, a Sume trending search is billed per search, so the same discipline applies in dollars rather than calls: decide the query first, keep the response, and do not repeat it. Check the pricing note on that page for the current rate.