Pricing

from $4.00 / 1,000 result (article, topic or profile)s

Try for free

Go to Apify Store

Toutiao News Scraper – 今日头条 China News, Hot Board & Comments

Try for free

Toutiao scraper API (今日头条, ByteDance news app), no login: Toutiao hot board/hot list = Chinese hot search, trending topics (头条热榜), Toutiao news & keyword search (头条新闻), article details + full text, channel feeds, Toutiao comments, replies, authors, followers, video URLs. $0.004/row, $0.0015/comment.

Pricing

from $4.00 / 1,000 result (article, topic or profile)s

Rating

0.0

(0)

Developer

AsiaData

Actor stats

Bookmarked

Total users

Monthly active users

22 days ago

Last modified

Categories

News

Social media

Security

Operation (hot board, news search, keyword research, article full text, comments, user articles, feed…)

operation

Optional

auto (default): urls/ids -> articleDetails, users -> userArticles, keywords -> search, channels -> feed, nothing -> hotBoard. hotBoard = 头条热榜 top 50. search = keyword news. keywordResearch = keyword -> top articles + full text + comments + free insights row. feed = channel news. articleDetails = full text. comments / replies. userProfile / userArticles. related. capabilities = free manifest. Use these exact names (Apify rejects other values before the run starts).

Type:string

Default:auto

Options:

autohotBoardsearchkeywordResearchfeedarticleDetailscommentsrepliesuserProfileuserArticlesrelatedcapabilities

Toutiao keyword search (news search keywords)

keywords

Optional

Keywords for operation search (头条搜索), one per item. Chinese works best; English names are auto-translated with expandQuery (BYD -> 比亚迪, Xiaomi -> 小米). Each row shows queryUsed. A Toutiao URL placed here is treated as a URL. Example: ["比亚迪", "artificial intelligence"].

Type:array

Toutiao article URLs (full text, comments, related)

urls

Optional

Toutiao article/video/微头条 URLs for articleDetails, comments or related (toutiao.com/article/…, /group/…, /w/…, /video/…, m.toutiao.com/i…/), or author profile URLs (toutiao.com/c/user//) for userProfile/userArticles. Invalid ones return an INVALID_URL error row (free). Example: ["https://www.toutiao.com/article/7694967967913607743/"].

Type:array

Toutiao article or comment IDs

ids

Optional

Numeric ids: article ids (the id field of article rows) for articleDetails/comments/related, or comment ids (commentId of comment rows) for operation replies. Example: ["7694967967913607743"].

Type:array

Toutiao user IDs or profile URLs (user articles, followers)

users

Optional

Toutiao author ids (authorId of any article row), https://www.toutiao.com/c/user// or /c/user/token// profile URLs (resolved automatically), for userProfile (incl. follower count) or userArticles. Example: ["1755235951852024"].

Type:array

Toutiao channel feeds

channels

Optional

Channel ids for operation feed (latest articles): news_hot=热点, news_tech=科技, news_finance=财经, news_world=国际, news_entertainment=娱乐, news_sports=体育, news_military=军事, all=推荐, news_game=游戏, news_health=健康, news_edu=教育, news_house=房产, news_history=历史, news_culture=文化, news_travel=旅游, news_food=美食, news_fashion=时尚, news_baby=育儿, news_regimen=养生, news_essay=美文, news_story=故事, news_agriculture=三农, news_astrology=星座, news_comic=动漫. Names like tech/科技 also work. Example: ["news_tech"].

Type:string[]

Max billable rows (total)

maxItems

Optional

Maximum billable rows for the whole run (articles, topics, profiles, comments, replies all count). Default 20; keywordResearch default = keywords x maxArticles x (1 + maxCommentsPerArticle). Range 1-10000. Free rows (error, summary, plan, insights) never count.

Type:integer

Max rows per input

maxItemsPerInput

Optional

Cap per keyword, channel, article (comments) or user. Default: maxItems split evenly between inputs. Max 1000.

Type:integer

Keyword research: articles per keyword

maxArticles

Optional

keywordResearch: how many top articles per keyword (after filters and sortBy) are opened for full text and comments. Default 10, range 1-100 (larger values are capped).

Type:integer

Keyword research: comments per article

maxCommentsPerArticle

Optional

keywordResearch: comments returned per article (Toutiao hot order), billed per comment. Default 20, range 0-500; 0 = articles only. Replies: repliesPerComment. Articles with fewer comments return what exists.

Type:integer

Include: full text, images, video URLs, author, comments

include

Optional

Extra data per article row. fullText = open each article: full text, images, exact time, all counts (same price). images = all image URLs. video = MP4 URL (expires ~1 h). author = nested profile (free). comments = nested top comments (billed per comment). insights = free heuristic analysis row. metrics / trust = add engagement metrics / trust fields in lean mode. Example: ["fullText"].

Type:string[]

Return only these fields

fields

Optional

Projection: keep only these output fields per data row (type is always kept). Empty = the default set of the outputMode. Names: id, url, title, abstract, text, source, authorName, authorId, publishedAt, readCount, likeCount, commentCount, shareCount, videoPlayCount, hotValue, searchQuery, commentId, replyCount, name, … (full list: operation capabilities). Example: ["title","url","publishedAt","readCount"].

Type:array

Output mode

outputMode

Optional

compact (default) = flat core fields, null when unknown. lean = token-efficient for LLM agents: only type, id and the most useful fields, no null/empty/false keys, operation/input/fetchedAt only once in the summary row, dates as 2026-10-10T10:02Z (UTC), text cut to maxTextChars (300). full = every field plus v0.x names (mode, query, groupId, scrapedAt).

Type:string

Default:compact

Options:

compactleanfull

Sort by

sortBy

Optional

Order within each input. relevance = Toutiao's order (default). newest = publish time. mostRead / mostLiked / mostCommented = by count. Non-default sorting scans up to 3x more candidates first.

Type:string

Default:relevance

Options:

relevancenewestmostReadmostLikedmostCommented

Published from

dateFrom

Optional

Keep only items published on/after this time. Formats: 2026-10-01 (Beijing time), ISO date-time, unix seconds, or relative 24h / 7d / 4w. Applied before billing. Example: "7d".

Type:string

Published until

dateTo

Optional

Keep only items published on/before this time. Same formats as dateFrom.

Type:string

Min reads

minReads

Optional

Keep only items with at least this many reads (readCount). Feed items without a read count need include ["fullText"]. Applied before billing.

Type:integer

Max cost (USD)

maxCostUsd

Optional

Stop before the run's charges would exceed this amount (e.g. 0.05). The run then ends SUCCEEDED with the rows so far and a COST_LIMIT note. Works together with Apify's max cost per run.

Type:number

Dry run (free plan)

dryRun

Optional

true = fetch nothing and bill nothing: returns one free plan row per input (detected operation, queryUsed, estimated rows, cost and seconds) plus a summary. Use it to check an input before paying.

Type:boolean

Default:false

Match terms (comments)

matchTerms

Optional

Keep only comments/replies whose text contains any of these words (case-insensitive); insights counts them per word. Applied after each page is fetched: lowers charges, not requests. Works for keywordResearch, comments, replies. Filter articles with textContains. Example: ["电池","续航"].

Type:array

Translate English keywords

expandQuery

Optional

Latin-script keywords are mapped to Chinese with a curated dictionary of ~950 brands, companies, people, products and news topics (BYD -> 比亚迪, Elon Musk -> 马斯克, Federal Reserve -> 美联储); unknown words stay as typed. Rows show queryUsed. false = search exactly as typed.

Type:boolean

Default:true

Snapshot mode: track metric changes

trackMetrics

Optional

true = compare with the previous run that used the same deltaKey: rows get snapshotAt, previousSnapshotAt, isNewSinceLastSnapshot and deltas (readsDelta, likesDelta, commentsDelta, sharesDelta, engagementRateDelta; hot topics hotValueDelta, rankDelta). Compact metrics only, kept 30 days in your account.

Type:boolean

Default:false

Snapshot key (trackMetrics)

deltaKey

Optional

Runs with the same deltaKey share one metrics snapshot (trackMetrics). One key per tracker, e.g. "hot-board" or "byd-news". Letters, digits, dashes. Default "default".

Type:string

Insights row (free, heuristic)

insights

Optional

true = add one free type=insights row: top terms, lexicon sentiment (labelled heuristic), engagement medians, top articles/comments, sources, commenter IP regions. Default: on for keywordResearch, off otherwise (false turns it off). No LLM.

Type:boolean

Replies per comment

repliesPerComment

Optional

Operations comments and keywordResearch: also return up to this many replies per comment as type=reply rows (billed as comment events). Default 0, max 100.

Type:integer

Default:0

Min likes

minLikes

Optional

Keep only items/comments with at least this many likes (likeCount). Applied before billing.

Type:integer

Min comments

minComments

Optional

Keep only items with at least this many comments (commentCount; for comments: replyCount). Applied before billing.

Type:integer

Text contains (any)

textContains

Optional

Keep only rows whose title, abstract or text contains at least one of these words (case-insensitive), checked before billing. Works for search, feed, userArticles, related, comments (comment text) and hotBoard (title). Example: ["比亚迪","BYD"].

Type:array

Comments per article (include comments)

commentsPerItem

Optional

With include ["comments"]: top comments nested in each article row. Default 5, max 100. Each comment is billed as a comment event.

Type:integer

Default:5

Lean format

leanFormat

Optional

Only with outputMode lean. rows (default) = normal dataset rows. table = also write key-value record LEAN_TABLE: one tab-separated table per row type with the header once (fewest tokens), copied into the summary row when under 20 KB. Rows are still pushed and billed the same; the table is a free extra.

Type:string

Default:rows

Options:

rowstable

Max text characters

maxTextChars

Optional

Cut text, abstract, description and comment text to this many characters (ends with …, row gets truncated: true). Default 300 in lean mode, no cut in compact/full. 0 = never cut; range 0–100000. Example: 1000.

Type:integer

Minimum:0

Maximum:100000

Content type

contentType

Optional

List operations: keep only one kind. any (default), article, video, micro-post (微头条). Videos come from feed, userArticles and related; keyword search returns articles only.

Type:string

Default:any

Options:

anyarticlevideomicro-post

Author tab (userArticles)

profileTab

Optional

Which part of an author's page userArticles reads. articles (default) = 头条号 articles, newest first, reads as rounded numbers. posts = micro-posts (微头条) with exact time, reads, likes, comments. all = both merged.

Type:string

Default:articles

Options:

articlespostsall

Search tab

searchType

Optional

search only. news (default) = News (资讯) tab, continuing in the All tab when it runs out. all = All (综合) tab only.

Type:string

Default:news

Options:

newsall

Include pinned hot-board item

includePinned

Optional

hotBoard only: also return the pinned official item above the ranked list (rank 0).

Type:boolean

Default:false

Include article HTML

includeHtml

Optional

articleDetails / include fullText: add raw article HTML as contentHtml.

Type:boolean

Default:false

De-duplicate

dedupe

Optional

Drop rows already returned earlier in the run (same type + id), e.g. when two keywords find the same article.

Type:boolean

Default:true

Only new rows (monitoring)

onlyNew

Optional

true = skip (and do not bill) rows already returned by earlier runs that used the same memoryName, e.g. a scheduled Hot Board or feed run every hour returns only new topics/articles. Seen ids are kept 30 days in named key-value store toutiao-seen- in your account.

Type:boolean

Default:false

Memory name (onlyNew)

memoryName

Optional

Runs with the same memoryName share one memory of seen rows (onlyNew). Use one name per monitor, e.g. "tech-feed" or "hot-board". Letters, digits and dashes.

Type:string

Default:default

Proxy configuration

proxyConfiguration

Optional

Apify Proxy (default, recommended): sessions rotate automatically on rate limits or search captchas.

Type:object

Default:

{
  "useApifyProxy": true
}

Max retries per request

maxRequestRetries

Optional

Retries with a fresh session/IP on rate limits, captchas or server errors (0-10).

Type:integer

Default:4

Delay between requests (ms)

requestDelayMs

Optional

Minimum delay between request starts to the same Toutiao host, to stay polite (0-10000). Default 100.

Type:integer

Default:100

Advanced News Scraper

dorcy/advanced-news-scraper

Extract the latest news articles with custom search queries, providing all the information, including article titles, sources, publication dates, full article text, and an AI-generated summary.

Dorcy Shema

250

Toutiao Scraper 📰

natanielsantos/toutiao-scraper

Easily scrape articles and videos from Toutiao (头条), a popular Chinese news and information content platform.

Nataniel Santos

372

Toutiao Scraper - Chinese News & Article Data API

sian.agency/toutiao-scraper

Scrape Toutiao (Jinri Toutiao) at scale: keyword-search Chinese news articles, extract full article content, engagement metrics (reads, comments, likes) and creator profiles. Export clean structured JSON for media monitoring, content intelligence and ByteDance research.

SIÁN OÜ

Yandex News Scraper

searchapi/yandex-news-scraper

Scrapes news articles from Yandex Search news results. Extracts headline, summary, source, publication date, author, thumbnail, and more.

Search API

Fast Google News Scraper

aymorato/fast-google-news-scraper

Extract details from Google News articles, such as images, titles, links, and other relevant information.

Alwin Morato

206

Medium Posts Search Scraper

easyapi/medium-posts-search-scraper

A powerful scraper that extracts comprehensive article data from Medium's search results. Get detailed information about articles, authors, and engagement metrics. Perfect for content research, trend analysis, and tracking popular writers and publications. 🔍📊

EasyApi

Google News All-In-One Api

romy/google-news-all-in-one-api

Real-time, unofficial Google News data as a REST API — personalized feeds, full-text search, regional editions, autocomplete, and category browsing. No scraping, no HTML parsing: this talks directly to the same private API the official Google News Android app uses, and returns clean JSON

Romy

今日头条 API | Toutiao Search, Comments & User Posts

socialdatax/socialdatax-toutiao-data-api

社媒数据助手 SocialDataX 提供只读今日头条数据：内容与用户搜索、详情、评论与回复、资料、作品、热榜和搜索建议。 Search Toutiao content and users; get details, comments, replies, profiles, creator posts, trending topics and search suggestions. 导出 JSON、CSV、Excel；无需配置 SocialDataX API Key。

SocialDataX

Google News Scraper — Headlines, Sources & Article URLs

scrapeify/google-news-scraper

Google News scraper for headlines and article URLs. Collect up to 2,000 deduplicated articles with titles, sources, dates, and snippets. Export CSV or JSON. No API key. Apify AI, MCP, and Cursor ready.

Scrapeify

Brave News Scraper

searchapi/brave-news-scraper

Searches Brave News and extracts rich, source-backed listing data including article URLs, publishers, snippets, images, publication times, displayed flags, and search context.

Search API