There Is No Official TikTok LIVE API. Here Is How to Get the Data Anyway

bash
curl "https://rngrow.com/api/v1/creators/resolve?username=somehandle" \
  -H "X-API-Key: rng_live_YOUR_KEY"

{ "found": true, "creator": { "username": "somehandle", "follower_count": 152300, "live": true } }

If you build your own checker, test it against a stream that just ended. That is the case that bites.

Trap 2: gift streaks double-count diamonds

TikTok gift combos report running totals while they animate. A naive recorder that logs every gift event will count the same diamonds several times, and every number downstream inflates. Correct accounting has to track deltas per streak, not raw events. Ask any data provider how they handle this; if the answer is vague, their earnings numbers are wrong.

Trap 3: freshness has to be in the response

LIVE data decays in minutes. A creator who was recruitable this morning may have signed with an agency by lunch. The only honest design puts verification timestamps on every record, so the consumer can decide what is fresh enough:

{
  "username": "creator_handle",
  "follower_count": 2126,
  "checked_at": "2026-08-29T06:54:14Z"
}

Ranking boards should also report their own age, and time-sensitive flags should expire: on our boards, the "was live at capture" flag is withheld once a board is older than 30 minutes, because at that point it is a lie.

Trap 4: offset pagination on a live table

The available-creators pool gains rows every second. Offset pagination against a table like that skips and duplicates rows between pages. Keyset (cursor) pagination on (checked_at, id) is stable:

import { Rngrow } from "@rngrow/sdk";

const rng = new Rngrow(process.env.RNG_API_KEY!);

for await (const creator of rng.creators.iterate({ country: "RO" })) {
  console.log(creator.username, creator.checked_at);
}

The iterator above walks the whole pool with cursors under the hood. The SDK is open source (github.com/steveantdev/rngrow-sdk, npm install @rngrow/sdk), zero dependencies, Node 18+.

What is actually available as data

More than most people expect. Beyond the creator pool, TikTok LIVE has competitive league divisions (ranked boards of 99 creators each), daily diamond leaderboards per country, per-game rankings for gaming streamers, and the gifting side: which spenders fund which creators. All of it can be structured:

# A full league division as JSON
curl "https://rngrow.com/api/v1/leagues/board?country=RO&division=A1" \
  -H "X-API-Key: rng_live_YOUR_KEY"

# The market's top gift senders, ranked by the last 7 days
curl "https://rngrow.com/api/v1/gifters?country=RO&sort=7d" \
  -H "X-API-Key: rng_live_YOUR_KEY"

Use cases I have seen work well: syncing recruitable creators into a CRM on a schedule, alerting a recruiting team when a ranked creator's diamonds spike (the single best moment to reach out), and qualifying leads by proven earnings instead of follower counts, which correlate with LIVE revenue far less than you would think.

Takeaways

- There is no official TikTok LIVE API, so all LIVE data is independently collected. Judge any source by whether you can verify its freshness from the response itself.
- The failure modes are silent: lingering room ids, double-counted gift streaks, stale "live" flags, offset pagination on moving tables. Design against them explicitly.
- If the data is a means to an end for you, consuming an API beats operating a pipeline. If you want to see what a finished surface looks like, our docs are at rngrow.com/api-docs and the overview is at rngrow.com/tiktok-live-api.

Disclosure: I built the API referenced here. The engineering traps apply no matter whose data you use, including your own.