TikTok Research API rate limit: max_count 100 vs Sume's 24
The Research API returns up to 100 videos per call (max_count). Sume's tiktok_search_keyword returns at most 24 items over 3 feed pages per call.

On the TikTok Research API Query Videos page, max_count defaults to 20 and tops out at 100 videos per response. Sume's tiktok_search_keyword route is a different surface: it returns at most 24 items and reads at most 3 feed pages per call, with no application step on the Sume side.
TikTok facts are from its Query Videos page (updated September 1, 2026); Sume facts from the OpenAPI document served with the API reference. Both read 2026-09-30.
What does the Research API cap per call?
The page defines max_count as the number of videos in the response: default 20, max 100. It also supports query operators (EQ, IN, GT, GTE, LT, LTE) over fields such as create_date, username and hashtag_name. The page itself does not state a request rate limit. A secondary search summary mentioned 1,000 requests and 100,000 records per day, but I did not confirm that on TikTok's page, so check TikTok's rate-limit documentation for your access.
What does Sume cap per call?
| Parameter | TikTok Research API | Sume tiktok_search_keyword |
|---|---|---|
| Items per call | max_count default 20, max 100 | limit 1 to 24, default 10 |
| Pages | Not a request field in the snapshot | max_pages 1 to 3, default 3 |
| Query | Operators over fields | query, 1 to 256 characters |
| Charge | Not on the page | No Sume credit charge in v1; upstream requests consume vendor credits |
What if I need more than 24 results?
Page with cursor (an integer on this route) across calls, and narrow with date_posted or region so each call returns something distinct. The route is bounded by design: the description says "At most 24 returned items and 3 feed pages." If you need 100 research records in one response, that is the Research API's access path, not Sume's.
curl -X POST https://api.sume.com/v1/scrapecreators/tiktok_search_keyword \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-d '{"query":"ugc skincare","limit":24,"max_pages":3}'What happens on a slow call?
Sync waits up to 30 seconds by default. A 202 means continue with jobs_wait then jobs_result using the request_id. The route is public data only: no login, stories, DMs, private accounts or automatic import, and it returns media candidates, not Sume-hosted media. Over MCP the same read is crawl_find with platform: "tiktok" and kind: "videos"; see MCP tools and gates. For trend ranking use the trending videos search API instead.
Sources
Related posts
More in Developers
- Join voiceover clips inside one timeline render with audio.parts
Timeline 1.0 takes up to 20 gapless audio.parts slices with source_in and duration. Skip the timeline-audio job when the join is only for one render.
- Timeline 1.0 limits: 200 slots, 1800 s, 20 audio parts, 8 fades
The numbers that cap one Timeline 1.0 render: 1 to 200 video slots, 1 to 1800 s of audio, 20 audio parts, 12-slot single strategy and 8 chained fades.
- Timeline render warnings: padded or looped short sources explained
A Timeline 1.0 slot longer than its source renders with a soft warning, not a failure, and /plan cannot predict it. Read warnings[] and probe clips first.
- Token bucket vs fixed window: what a reset header means
Anthropic says its limits replenish continuously; Sume's docs call ratelimit-reset the seconds until the window resets. How to pace a client for each.
Written by Sume