Douyin API Python

Call Douyin (the Chinese TikTok) from Python: search videos by keyword with likes, comments, shares and saves, then pull the comments of a video with likes, reply counts and the commenter's region. You do not need a Douyin developer account or a mainland China network: each call is one POST request to the EveryInfra API.

Also covers: Douyin API alternative, Douyin scraper.

Quick start

git clone https://github.com/everyinfra/douyin-api-python.git
cd douyin-api-python
pip install -r requirements.txt
export EVERYINFRA_API_KEY=your_key   # get one at everyinfra.com
python main.py search 咖啡 --out videos.csv
python main.py comments https://www.douyin.com/video/<id> --out comments.csv

Get an API key: everyinfra.com/en/docs. Without --out, the script prints the first three rows as JSON.

Price

Call	Price per 1,000 calls	Rows per call
douyin/search	$8.33	10 in our test
douyin/comments	$0.56	20 in our test

• You pay per call, not per month. Failed calls and empty results are not charged.
• New accounts can claim $1 of trial credit: create an account, star everyinfra-agent-plugins and seo-skills, then claim it in the console. It is valid for 24 hours and needs no card. See pricing.

What you get

search

Field	Meaning
next_page_token	Page token (search / user_posts / comments / comment_replies / billboard_video_hot / billboard_low_fan_explosion / billboard_high_completion / billboard_account_search / music_search): pass it back unchanged as page_token to fetch the next page, keeping the other parameters the same; valid for 24 hours. Every row of a page carries the same value, which also appears at the top level of the response. Absent on the last page. Comment pages walk all top-level comments page by page.
id	The record's unique ID on the platform. Unique only within one platform; IDs can repeat across platforms.
url	Link to the record on the platform.
text	Body text with HTML removed, returned in full and never cut short (on a few platforms the list results themselves carry only an excerpt; the platform's field notes say so).
media_type	Media type: video, or image for image posts.
video_url	Video playback URL (signed, expires; download promptly if you need to keep it). Null for image posts or when the source omits it.
like_count	Likes. null means the platform or this capability does not disclose it; it is not 0.
comment_count	Comments. null means not disclosed, not 0.
share_count	Shares. null means not disclosed, not 0.
collect_count	Saves or favorites. null means not disclosed, not 0.
view_count	Play count. Douyin does not publish play counts, so this field never carries real plays; judge reach by like_count, comment_count and share_count.
author_name	Author's display name.
author_id	Author's unique ID on the platform.
author_douyin_id	The author's Douyin ID.
author_uid	Numeric user ID of the author (on comments, of the commenter). author_id is the secUid; both identify the same person.
author_bio	The author's profile bio (search / user_posts / video rows).
author_avatar_url	Author avatar image URL: the creator on search, user_posts and video rows, the commenter on comments rows.
author_follower_count	Author's follower count. null means not disclosed.
cover_image_url	Video or image-post cover thumbnail URL (search / user_posts rows).
video_width	Video frame width in pixels.
video_height	Video frame height in pixels.
duration_seconds	Duration in seconds. null for content that is not video or audio.
region	Region.
hashtags	Hashtags in the text, in order of appearance; absent when there are none.
posted_at	Publish time in ISO 8601, UTC (for example 2026-08-07T12:34:56+00:00). Non-standard strings from the platform are passed through as is, so parse defensively. null means the platform does not disclose it.
platform	Source platform identifier, the same as platform in the request (for example xiaohongshu, tiktok).
image_urls	Array of image links in the platform's display order; absent when there are no images.
duration_sec	Video length in seconds.
author_is_verified	Whether the author is verified (search / user_posts / video rows).
author_following_count	How many accounts the author follows (search / user_posts / video rows).
author_like_count	Total likes the author's profile has received (search / user_posts / video / transcript).
author_video_count	Total number of posts on the author's profile (search / user_posts / video rows).
author_ip_location	The author's IP location, which Douyin displays on every profile.
title	Title. null when the platform has no titles (for example text-only posts).
music	Background track: id, title, author, is_original (original sound or not), duration_sec, cover_url and play_url (audio link, may expire). Absent when there is no music information.
download_count	How many times the post was downloaded.
forward_count	How many times the post was forwarded.
is_pinned	Whether the item is pinned (post rows: pinned on the author's profile; trending: pinned on the board).
is_ad	Whether the post is an ad.
video_tags	Content category tags Douyin assigned to the post (for example food or store visit).
video_resolution	Video quality tier as Douyin labels it (for example 720p).
animated_cover_url	Animated cover link.
city	City name in the platform's language, not normalized.
location	Location description passed through as is; the format varies by platform.

Also returned, named as on the platform: hashtag_ids, search_rank, city_code, text_language, video_bitrates, download_url, has_watermark, video_url_expires_at, permissions, author_gender, author_age.

Full parameter list: Douyin search API.

comments

Field	Meaning
next_page_token	Page token (search / user_posts / comments / comment_replies / billboard_video_hot / billboard_low_fan_explosion / billboard_high_completion / billboard_account_search / music_search): pass it back unchanged as page_token to fetch the next page, keeping the other parameters the same; valid for 24 hours. Every row of a page carries the same value, which also appears at the top level of the response. Absent on the last page. Comment pages walk all top-level comments page by page.
id	The record's unique ID on the platform. Unique only within one platform; IDs can repeat across platforms.
text	Body text with HTML removed, returned in full and never cut short (on a few platforms the list results themselves carry only an excerpt; the platform's field notes say so).
like_count	Likes. null means the platform or this capability does not disclose it; it is not 0.
reply_count	Replies, usually replies to a comment. null means not disclosed.
author_name	Author's display name.
author_id	Author's unique ID on the platform.
ip_location	IP region Douyin displays (set by the platform, not entered by the user).
is_hot	Whether the comment carries Douyin's hot-comment badge (comments).
is_reply	Whether the row is a reply.
reply_to_id	ID of the top-level comment this reply sits under: on comments reply rows, the comment it hangs from; on every comment_replies row, the comment you asked about.
posted_at	Publish time in ISO 8601, UTC (for example 2026-08-07T12:34:56+00:00). Non-standard strings from the platform are passed through as is, so parse defensively. null means the platform does not disclose it.
video_id	ID of the video the comment belongs to (comments).
platform	Source platform identifier, the same as platform in the request (for example xiaohongshu, tiktok).
liked_by_author	Whether the video's author liked the comment; comments the author liked tend to carry more weight.
author_avatar_url	Author avatar image URL: the creator on search, user_posts and video rows, the commenter on comments rows.
author_uid	Numeric user ID of the author (on comments, of the commenter). author_id is the secUid; both identify the same person.
images	Array of image links in the platform's display order; absent when there are no images.
mentioned_users	People @-mentioned in the comment, each with user_id (secUid), uid (numeric ID) and nickname. Absent when nobody is mentioned.

Full parameter list: Douyin comments API.

Tested

We ran both commands against the live API before publishing this repository (times in Singapore time, UTC+8).

Command	Run at (SGT)	Rows	Charged
search 咖啡	2026-11-08 03:37	10	$0.008333
comments https://www.douyin.com/video/<id>	2026-11-08 03:37	20	$0.000556

First row of the search result (long text cut, author fields left out of this sample):

{
  "id": "7510430154377334075",
  "url": "https://www.douyin.com/video/7510430154377334075",
  "text": "咖啡不再傻傻分不清 #咖啡#中国成都国际非遗节 #百young非遗计划",
  "media_type": "video",
  "like_count": 2153372,
  "comment_count": 42444,
  "share_count": 961296,
  "collect_count": 448215,
  "region": "CN",
  "posted_at": "2025-05-31T02:24:50+00:00",
  "platform": "douyin",
  "title": "咖啡不再傻傻分不清",
  "is_pinned": false,
  "video_width": 1080,
  "video_height": 1920,
  "duration_sec": 37,
  "hashtags": [
    "咖啡",
    "中国成都国际非遗节",
    "百young非遗计划"
  ],
  "duration_seconds": 37
}

How it works

Every call is one request. everyinfra.py wraps it in about 60 lines of requests:

curl -X POST https://api.everyinfra.com/api/v1/social \
  -H "Authorization: Bearer $EVERYINFRA_API_KEY" \
  -H 'Content-Type: application/json' \
  -d '{"platform": "douyin", "action": "search", "params": {"keyword": "咖啡"}}'

• Slow actions (Instagram user_posts, for example) answer with a job_id. The script polls GET /api/v1/jobs/{job_id} every 5 seconds; polling is free.
• Errors come back as JSON with a code, a message and whether you were charged. A missing parameter returns 422 before any charge.
• The same key also works over MCP for Claude, Cursor and other AI agents: docs.

More

• Douyin API on EveryInfra: every Douyin action, with prices and live health
• Other examples: google-maps-scraper · instagram-scraper-python · tiktok-trends-api · tiktok-scraper-python

License

MIT