Temu Scraper: Products, Prices, Variants & Reviews

Paste Temu product links or the Temu home page link, and get Temu products as clean rows with price, currency, sales, images and a product link. Product links also add rating, review count, variants, specifications and public reviews.

Use it for Temu price checks, dropshipping research and tracking the products you already know, in the US market. This version has no keyword search and no category pages. The prefilled home feed with 10 products finished in about 3 to 7 seconds in our tests.

Table of contents

• What it does
• How it works
• Quick start
• Use cases
• Input
• Output
• Pricing
• Limits
• API, export and integrations
• Troubleshooting
• FAQ
• Related Actors
• Support

What it does

This Actor reads public Temu pages for the US market and saves one row for each product. It reads two things: the Temu home feed and single product pages. Prices are in USD. You do not need a Temu account.

A home feed row has the title, price, image, link and usually sales and the original price. Temu does not send ratings or review counts in the feed, so those fields are null there. A product page adds the rating, review count, variants, specifications, the description and the public reviews shown on that page.

Good fit

• Read variants, ratings and public review text for a list of product links.
• Check the prices of products you already know, on a schedule.
• See what Temu shows on its US home feed right now, with price and sales.

Not a fit

• Finding products by keyword. This version has no search.
• Exporting a Temu category. This version does not read category pages.

👉 Try it now: open the input form

How it works

1. You give product links, product IDs or the Temu home page link.
2. The Actor reads those public pages, applies your filters and removes duplicate products.
3. It saves one row per product in the Dataset. A short report of the run goes to the Key-value store.

👉 Try it now: open the input form

Quick start

1. Open the Input tab. Keep the home page link in Temu links, or replace it with your product links (box 1 in the picture).

2. Keep Maximum products at 10 for the first run (box 2).

3. Click Start (box 3). The picture shows the owner view, where this button reads Save & start.

4. Open the Output tab and check the Products table: title and price first (box 4).

   

Expected result: 10 rows, each with a title, a USD price, an image and a Temu link. On feed rows rating and reviewCount are null. Use a product link or dataLevel: full when you need them.

👉 Try it now: open the input form

Use cases

Buyer job	Input path	Output entity	Key fields	Next integration
Variant and review check	productUrls	Full product row	skus, reviews, specifications, minPrice, maxPrice	JSON or API
Price tracking for known products	productUrls, mode: monitor, stateName	Changed product row	changeType, changedFields, price	Schedule and webhook
Home feed price and sales export	startUrls with the home page, maxItems	Product row	title, price, discountPercent, sales, url	CSV or Google Sheets
Home feed with ratings	startUrls with the home page, dataLevel: full	Full product row	rating, reviewCount, skus, sales	Excel or API

👉 Try it now: open the input form

Input

Input fields

Field	Plain name	Type	Required	Default / prefill	Allowed values	Effect	Example	Recommendation
startUrls	Temu links	array of strings	one of the two	prefill ["https://www.temu.com/"]	up to 1000 product or home links	The home link reads the feed, product links read those products	["https://www.temu.com/"]	Keep the home link for a first try
productUrls	Product URLs or IDs	array of strings	one of the two	none	up to 1000 Temu links or numeric IDs	Reads full product pages	["606102494273968"]	Remove the home link if you want only these products
country	Country	string	no	US	US	Sets the market	US	Keep it. Other markets are not supported
maxItems	Maximum products	integer	no	10	1–100000	Stops after this many saved rows across all inputs	200	Raise it after the first run
mode	Mode	string	no	auto	auto, products, home, monitor	Chooses which inputs run	monitor	Keep auto unless you want one input type
dataLevel	Data level	string	no	auto	auto, listing, full	auto and listing save rows from the feed. full also opens each product page	full	Use full only when you need ratings, variants or reviews
minPrice	Minimum price	number	no	none	0 or more	Keeps products at or above this price	5	Use for price research
maxPrice	Maximum price	number	no	none	0 or more	Keeps products at or below this price	20	Use for price research
minRating	Minimum rating	number	no	none	0–5	Skips lower ratings and products with no rating	4.5	Feed rows have no rating. Use it with product pages
minReviews	Minimum review count	number	no	none	0 or more	Skips products with fewer reviews	100	Feed rows have no review count. Use it with product pages
minSold	Minimum sold count	number	no	none	0 or more	Skips products with fewer sales. 1K+ counts as 1000	1000	Works on the feed
minDiscount	Minimum discount (%)	number	no	none	0–100	Skips smaller or missing discounts	30	Works on the feed when Temu shows a discount
includeKeywords	Title must contain	array of strings	no	none	up to 20	Keeps titles with every entered word. It filters rows and does not search Temu	["waterproof"]	Use for a narrow list
excludeKeywords	Title must not contain	array of strings	no	none	up to 20	Skips titles with any entered word	["connector"]	Removes accessories
excludeAds	Exclude ads	boolean	no	false	true or false	Skips products marked as ads	true	Turn on for clean lists
excludeAdultProducts	Exclude adult products	boolean	no	false	true or false	Skips products marked as adult	true	Turn on for shared reports
onlyNew	Only new products	boolean	no	false	true or false	Saves products not yet in this history	true	Use for scheduled runs
resetState	Reset history	boolean	no	false	true or false	Starts this history again	true	Use once, then turn off
stateName	History name	string	no	default	lowercase letters, numbers, hyphens, up to 40 characters	Keeps separate history per project	my-mugs-us	Use one name for each project
deliveryConnector	Send results to	string	no	none	your MCP connector	Posts a short summary after the run	Slack connector	Optional
deliveryTarget	Notion page or channel	string	no	none	Notion page link or Slack channel ID	Chooses where the summary goes	C0123456789	Optional

An empty input stops the run with an error and no catalog or product fee, because the Actor does not guess what you want. The same applies to an invalid value, such as a link from another site, a keyword, or a Temu search or category link. Duplicate products within a run are saved once.

Interacting settings

When	Input A	Input B	Effective behavior	Example	Cost/result effect
Both input types	mode: auto	home link and productUrls	Both run until maxItems	The prefilled home link stays next to your links	Remove the home link if you want only your products
One input type	mode: products	home link	Only product links run	The home link stays unused	No fee for skipped inputs
Product links	productUrls	dataLevel: listing	Product links always use full pages	One product ID	product-details fee applies
Full data	dataLevel: full	filters or onlyNew	A checked full page is charged even when the row is skipped	minPrice above all prices	The page stays in a PRODUCT_ID file
Filters on the feed	filters	maxItems	The feed reads about maxItems products and at least 60, then stops	minSold: 1000, 10 products	You can get fewer rows than maxItems
History	onlyNew or mode: monitor	stateName	The same name continues the history	my-mugs-us	Unchanged feed rows cost no product fee. Product pages are still charged
Fresh start	resetState	stateName	This history starts empty	Reset once	All rows count as new

Input examples

First run: home feed

{
  "startUrls": ["https://www.temu.com/"],
  "maxItems": 10
}

Expected behavior: the Actor reads the first feed page and saves 10 products. Expected output: product rows with title, price, currency, url, images, usually sales. Limit or cost note: about $0.049 on the Free plan.

Variants and reviews

{
  "productUrls": ["606102494273968"],
  "maxItems": 1
}

Expected behavior: the Actor opens the product page. Expected output: one row with skus, specifications, reviews, minPrice, maxPrice. Limit or cost note: one product-details fee. If Temu blocks the page, you pay no fee for it.

Track prices of known products

{
  "productUrls": ["606102494273968", "601102610144502"],
  "mode": "monitor",
  "maxItems": 2,
  "stateName": "my-products-us"
}

Expected behavior: the first run saves the products. The next runs save only new or changed ones. Expected output: rows with changeType and changedFields. Limit or cost note: each checked product page pays product-details, even with no changes.

Home feed with ratings and variants

{
  "startUrls": ["https://www.temu.com/"],
  "dataLevel": "full",
  "maxItems": 10
}

Expected behavior: the Actor reads the feed and opens the page of each product. Expected output: rows with rating, reviewCount, skus, reviews. A row stays partial if Temu blocks its page. Limit or cost note: one catalog fee plus one product-details fee per opened page, about $0.19 for 10 on the Free plan.

👉 Try it now: open the input form

Output

Output fields

Field	Meaning	Type	Can be empty?	Example / enum	Source / notes
goodsId	Temu product ID	string	no	606102494273968	From the page
title	Product title	string	conditional	1pc Vacuum Insulated Coffee Travel Mug	From the page
price	Current price as a number	number	conditional	6.61	In currency units
priceText	Price as Temu shows it	string	conditional	$6.61	Raw text
currency	Currency	string	conditional	USD	From the page
originalPrice	Price before discount	number	yes	35.99	Only when Temu shows it. It can differ between runs
discountPercent	Discount in percent	number	yes	79	Only when Temu shows it. It can differ between runs
minPrice / maxPrice	Lowest and highest price	number	yes	6.61 / 12.36	Differ when variants differ
sales	Sold count as a number	number	yes	28000	Rounded badges give a lower bound
salesText	Sales badge	string	yes	28K+ sold	Raw text
salesIsLowerBound	The sales number is rounded down	boolean	yes	true	True for badges such as 28K+
rating	Average rating out of 5	number	yes	4.8	Product pages only. null on feed rows
reviewCount	Number of reviews	number	yes	645	Product pages only. null on feed rows
url	Product link	string	conditional	https://www.temu.com/-g-606102494273968.html	Public page
images	Image links	array	yes	["https://img.kwcdn.com/..."]	Full pages may have more
videoUrl	Product video link	string	yes	null	When shown
skuId	Variant ID picked in the listing	string	yes	109023852495536	When shown
skus	Variants with ID, price, specifications and stock	array	yes	10 items on the example	Full pages
reviews	Public reviews on the page: text, rating, date, media	array	yes	4 items on the example	Full pages, no profile data
specifications	Product specifications	array	yes	material, size	Full pages
description	Product description	string	yes	text	Full pages
sellerId / sellerName / sellerUrl	Public shop ID, name and link	string	yes	634418231860874	When shown
categoryId / categoryPath	Category ID and names	string	yes	2407	Full pages, when shown
inStock	Stock status	boolean	yes	true	null means Temu did not report it
shippingText / returnPolicy	Shipping and return text	string	yes	text	Full pages, when shown
isAd / isAdultProduct	Temu marks it as an ad or adult product	boolean	yes	false	When shown
tags	Public labels	array	yes	[]	Listing badges
rank	Position in the feed before your filters	integer	yes	1	Feed rows
detailLevel	Row source	string	conditional	listing, full	Tells you what to expect
status	Row state	string	conditional	complete, partial	partial means the product page failed and the feed row stayed
changeType / changedFields	What changed in monitor mode	string / array	yes	changed, ["price"]	Monitor mode only
country, source, rowType	Market, source and row type	string	no	US, temu, product	Fixed values
sourceUrl, fetchedAt	Page read and time in UTC	string	conditional	2026-11-05T11:45:35.401Z	For your records

Output entities and surfaces

Situation	Dataset/API/KVS/live view	Result shape	User action
Normal item	Dataset, views Products, Product variants, Price and first variant, All fields	One row per product	Open the table or export it
Domain-empty result	Empty Dataset, RUN_SUMMARY with status not_found, no_matches or no_changes	No rows, run SUCCEEDED	Check the link, the filters or the history
Partial result	Dataset rows, status partial in RUN_SUMMARY, stopReason CHARGE_LIMIT, DEADLINE or ITEMS_FAILED	Saved rows stay, run SUCCEEDED	Read errors and unprocessedInputs, then run the rest again
Technical failure or retry	RUN_SUMMARY with stopReason SOURCE_ERROR or INVALID_INPUT, run FAILED	Rows saved before the stop stay	Try again later or fix the input

Full product pages are also kept as PRODUCT_<ID> files in the Key-value store.

Output example

{
  "goodsId": "606102494273968",
  "title": "1pc Vacuum Insulated Coffee Travel Mug, Leak-Proof Double Wall Stainless Steel Reusable Vacuum Cup",
  "price": 6.61,
  "currency": "USD",
  "originalPrice": null,
  "minPrice": 6.61,
  "maxPrice": 12.36,
  "sales": 28000,
  "salesText": "28K+ sold",
  "salesIsLowerBound": true,
  "rating": 4.8,
  "reviewCount": 645,
  "url": "https://www.temu.com/-g-606102494273968.html",
  "detailLevel": "full",
  "status": "complete"
}

This example is trimmed and contains one product from the "Variants and reviews" input. null means Temu did not show the value on that page. It does not mean zero. We do not promise that every product has every field.

Send results to Slack or Notion

Pick your connector in the optional deliveryConnector field. After the run saves its data, the Actor posts a short summary (up to 25 rows and a link to the full dataset) there. Put a Notion page link or a Slack channel ID in deliveryTarget: in Notion every run is added on top of that one page, so your workspace does not fill up with new pages. A delivery error never fails the run and costs nothing extra. Connect a service once in Apify Console → MCP connectors.

Use case → fields

Buyer job	Input path	Output entity	Key fields	Next integration
Variant check	productUrls	Full product row	skus, minPrice, maxPrice	JSON
Review check	productUrls	Full product row	reviews, rating, reviewCount	JSON
Change tracking	productUrls, mode: monitor	Changed product row	changeType, changedFields	Webhook
Feed export	home link	Product row	price, sales, discountPercent	CSV

👉 Try it now: open the input form

Pricing

Pricing unit

You pay per event. Platform usage is included in these prices.

Event	When it is charged	Free	Bronze	Silver	Gold and above
Actor start	Once per run, per GB of memory	$0.00005	$0.00005	$0.00005	$0.00005
Catalog check	Up to three home feed pages. Includes three basic products	$0.03	$0.03	$0.03	$0.03
Product	One saved feed row above the included three	$0.0027	$0.00243	$0.00216	$0.00189
Product details	One full product page	$0.0162	$0.01458	$0.01296	$0.01134

• The start fee grows with run memory, per GB. The default is 256 MB.
• A feed page holds about 20 products.
• A filter with no matches pays the catalog fee, the same as a feed run where no row is new. They pay for reading the feed, not for rows.
• A checked full product page is charged even if a filter or the history skips the row. You can still read it in its PRODUCT_<ID> file.
• A page that Temu blocks costs no catalog, product or details fee.
• If a run stops early, you pay only for the work done before the stop.
• Set Maximum cost per run (maxTotalChargeUsd in the API) to control the cost. At the limit, the run stops and keeps the saved rows. A run with no limit can spend up to your account balance, so always set one for large API runs.

Cost examples

Prices are for the Free plan.

Scenario	Input scope	Expected item/event count	Base charge	Optional charge	Stop/cap behavior
First run	Home feed, 10 products	1 catalog fee, 7 products	$0.04895	none	Under $1
Home feed	60 products	1 catalog fee, 57 products	$0.18395	none	Under $1
Whole feed	Asked for 300, got 238	4 catalog fees, 226 products	$0.73025	none	Under $1
Feed with full pages	10 products, dataLevel: full	1 catalog fee, 10 details	$0.19205	none	Under $1
One product page	1 product link	1 details	$0.01625	none	Under $1
20 product pages	20 product links	20 details	$0.32405	none	Under $1
No matches	Home feed, a filter skips all	1 catalog fee	$0.03005	none	No product fee
Monitor, 5 products, no changes	5 product links	5 details	$0.08105	none	Pages are charged, rows are not saved

Paid plans pay less for product rows, as the table above shows.

👉 Try it now: open the input form

Limits

• No keyword search and no category pages. Temu refuses these pages to anonymous visitors for hours at a time, and we did not find a stable way to read them without an account. A keyword, a search link or a category link stops the run with a clear error and no catalog or product fee.
• US market and USD only. Other Temu markets are not supported.
• The length of the home feed changes from run to run. It ended at 238 products when we asked for 300, and at 40 products when we asked for 60. Both runs finished as complete. maxItems of 100000 is an input limit, not a promise that Temu returns that many.
• The feed is not the whole catalog. It changes between runs: in one pair of runs a few minutes apart half of the 60 products were different, and in another pair only 1 product was the same. Temu picks what it shows, and you cannot choose a topic.
• Feed rows have no rating and no review count. Open product pages with dataLevel: full or use product links to get them.
• The price in a feed row can differ from the price on the product page. In one test a product was $12.24 in the feed and $15.97 on its page. Use product links when the exact page price matters.
• originalPrice and discountPercent can change or disappear between two runs a few minutes apart. Monitor mode does not count them as a change. It compares title, price, currency, sales, rating, review count and stock.
• Temu can also block single product pages. Each page gets up to four attempts. In our last three test batches 56 of 60, 38 of 43 and 41 of 43 product pages were saved, and the others were reported in errors. A blocked page costs no fee. Run those links again. After eight failed reads in a row, which is two product pages in a row with all attempts refused, the run stops to avoid wasting traffic and ends as FAILED with SOURCE_ERROR. Saved rows stay in the Dataset.
• If Temu refuses the feed in the middle of a large run, the run ends as partial. Saved rows stay in the Dataset, and unprocessedInputs shows what is missing.
• After three failed reads of a feed page the run stops with SOURCE_ERROR. Saved rows stay in the Dataset.
• The default run timeout is 300 seconds. Raise it for large runs: 20 product pages took 137 seconds, so plan for about 35 pages per run at the default.
• The history keeps up to 10000 products for each stateName.
• Monitor and onlyNew work best with product links. On the home feed a "new" row means new in your history, not new on Temu, because the feed changes on its own.
• The Actor reads public data only. It does not log in, solve CAPTCHAs or collect reviewer profiles or contact details.

👉 Try it now: open the input form

API, export and integrations

• API: start a run with the input JSON through the Apify API, then read rows from the run's Dataset. Use goodsId as a stable key.
• Export: download the Dataset as JSON, CSV, Excel, XML or HTML from the Output tab or the API.
• Scheduling: create a task with mode: monitor and product links and a fixed stateName, then add a schedule in Apify Console. Read changeType and changedFields in the new rows.
• Webhooks / no-code: use a webhook on the run finish event. Make, Zapier and n8n can read the Dataset through the Apify integrations.
• SDK / MCP / agent: call the Actor by cliqtomedia/temu-scraper from the Apify SDK or the MCP server below.

Use with AI agents (MCP)

MCP server URL with only this Actor: https://mcp.apify.com?tools=cliqtomedia/temu-scraper.

Smallest useful input:

{
  "productUrls": ["606102494273968"],
  "maxItems": 1
}

Key output fields:

• title: product title.
• price and currency: current price as a number and its currency.
• rating, reviewCount, skus, reviews: filled on product pages.
• url: public Temu product link.
• goodsId: stable product ID.
• detailLevel: listing or full, which tells if skus and reviews can be filled.

This input cost $0.01625 on the Free plan. An empty input is an error, and this Actor cannot search by keyword. Set maxTotalChargeUsd on each API run. The default timeout is 300 seconds. The US market is the only market.

👉 Try it now: open the input form

Troubleshooting

The run failed with INVALID_INPUT

Symptom: the run status is FAILED and stopReason in RUN_SUMMARY is INVALID_INPUT. Likely reason: the input is empty, or it has a keyword, a Temu search link, a category link or a link from another site. Check: read the error text in RUN_SUMMARY. Action: use product links, product IDs or the link https://www.temu.com/.

The run failed with SOURCE_ERROR

Symptom: the run status is FAILED and stopReason in RUN_SUMMARY is SOURCE_ERROR. Likely reason: Temu pages could not be read several times in a row: three feed reads of one page, or eight product page reads. Check: open errors and unprocessedInputs in RUN_SUMMARY. Rows saved before the stop stay in the Dataset. Action: try again later.

The run finished with fewer rows than I asked for

Symptom: status in RUN_SUMMARY is partial, or there are fewer rows than maxItems. Likely reason: the run reached the cost limit (CHARGE_LIMIT) or the time limit (DEADLINE), a product page failed (ITEMS_FAILED), a filter skipped rows, or the feed ended. Check: read stopReason, detailFailures and unprocessedInputs. Action: raise Maximum cost per run or the timeout, then run the rest again. The feed can end early: 238 products at most in our tests, and 40 in the shortest run.

My product links were read, and I also got random products

Symptom: the Dataset has products you did not ask for. Likely reason: the prefilled home link is still in Temu links. Check: look at sourceUrl in those rows. Action: remove the home link or set mode to products.

Rating and review columns are empty

Symptom: rating or reviewCount are null. Likely reason: the row came from the home feed. Temu does not send these values there. Check: look at detailLevel in the row. Action: use productUrls, or set dataLevel to full.

The Dataset is empty and the run succeeded

Symptom: 0 rows, and status is not_found, no_matches or no_changes. Likely reason: filters skip everything, or the history has no changes. Check: read scanned, filtered and unchanged in RUN_SUMMARY. Action: relax the filters, or set resetState to true.

The history warning appears

Symptom: the status message says the history store is unavailable, and RUN_SUMMARY has HISTORY_UNAVAILABLE. Likely reason: another Actor owns a store with the same name, or access was refused. Check: read warnings in RUN_SUMMARY. The rows are still saved. Action: use a new stateName.

👉 Try it now: open the input form

FAQ

Can it search Temu by keyword?

No. Temu refuses search and category pages to anonymous visitors for hours at a time, so this version does not offer them. If you need keyword search, this Actor is not the right tool. With product links it works without search.

Do I need a Temu account or cookies?

No. The Actor reads public pages. You do not add any login, cookie or key.

Is this the Temu API?

No. Temu did not make this Actor. You get your own API through Apify: start a run, then read the Dataset as JSON.

Do I need a browser extension or a tool to install?

No. It runs in Apify Console. You can also start it through the API.

Why does a run with no matches cost money?

The fee pays for reading the feed. A filter with no matches pays only the catalog fee and the start fee, with no product fee.

How fresh is the data?

Each row shows the read time in fetchedAt. Temu changes prices and product lists often, so two runs can differ.

Can I get all reviews of a product?

No. The Actor saves the public reviews shown on the product page, in reviews. reviewCount shows the total that Temu reports.

👉 Try it now: open the input form

Related Actors

• AliExpress Scraper: compare the same product on AliExpress.
• Taobao & 1688 Scraper: check supplier prices in China.
• Google Shopping Scraper: compare prices across stores.
• Shopify Store Products Export: export the catalog of a store you want to compare.
• Trendyol Scraper: read another marketplace with the same kind of rows.

👉 Try it now: open the input form

Support

For a bug, a question or a feature request, open the Issues tab of this Actor on Apify. Include the run ID, the input mode and the error text from RUN_SUMMARY. Never send tokens, cookies or billing data.

👉 Try it now: open the input form