Replicate AI Models Scraper

Replicate AI Models Scraper

Scrape AI model metadata from Replicate's Explore page or any collection, up to a million models per run. Every model comes with its name, description, run count, owner, and hardware details. No API key required. Export to CSV, JSON, Excel, or XML.

Replicate's official API needs a token and pagination logic for browsing models. This reads the public model feeds directly, filtered by collection slug, and returns each match in one fixed schema. You get a flat dataset of AI models without writing a single line of code.

Who uses it	What they scrape Replicate for
AI developers	Find trending open-source models for a new project.
Product managers	Monitor which model categories are growing fastest.
Market researchers	Track model popularity and run counts over time.
Content curators	Build a searchable directory of AI models by task.

What it does

This Actor collects AI model listings from Replicate by collection or from the main Explore feed, and returns each model as a flat row.

• 🧭 Explore feed: scrape featured, official, and trending models from the main Explore page.
• 📂 Collection scraping: target a specific collection like text-to-image, image-editing, or language-models.
• 🔢 Flexible volume: collect anywhere from a single model up to 1,000,000 per run.

Results export to CSV, JSON, Excel, or XML, or straight from the API.

What you can do with Replicate data

📈 Track model trends.

A market researcher scrapes the text-to-image collection weekly to see which new models are gaining runs and which are fading.

🧪 Find models for a prototype.

An AI developer scrapes the language-models collection to compare open-source LLMs by run count and hardware before integrating one.

📋 Build a model directory.

A content curator scrapes the full Explore feed daily to populate a searchable website of AI models with descriptions and owner names.

📊 Monitor the competitive landscape.

A product manager scrapes the image-editing collection to see which companies are publishing the most models and how their run counts compare.

Why choose this scraper

	What you get
No API key	Scrapes the public web feeds, no token or registration needed.
Fixed schema	Every model returns the same fields, ready for analysis.
Collection filter	Narrow results to a single task category like text-to-video.
High volume	Collect up to a million models in one long run.

How it compares

No other Store actor targets Replicate the same way, so the honest comparison is with the alternatives teams actually weigh.

	Replicate AI Models Scraper	Build it in-house	By hand
Setup	Run it now, zero config	Days of engineering	None, but hours per pull
When Replicate changes	Maintained for you	You fix it	You re-learn the page
Proxies, retries, anti-bot	Built in	Your problem	Browser only
Output	Fixed JSON schema, CSV/Excel export	Whatever you build	Copy-paste
Cost	Pay per result	Engineering time	Analyst hours

Configure the run

Drive the Actor from the Explore page or a collection slug, and set a maximum number of models to collect so only the newest or most relevant entries reach your dataset. The Input tab lists every parameter.

A first run with the defaults:

{
  "maxItems": 10
}

A larger pull:

{
  "maxItems": 200
}

Pricing

Pay-per-result: $0.0085 per result collected. You pay only for the results written to your dataset.

Results collected	Approximate cost
100 results	$0.85
1,000 results	$8.50
10,000 results	$85.00

New Apify accounts start with $5 in free credit.

Free users

Free-plan runs return up to 10 results as a preview. Upgrade your Apify plan to collect up to 1,000,000 results per run.

Run it

1. Create a free Apify account with $5 in credit.
2. Open the Replicate AI Models Scraper.
3. Set your inputs and any filters, then click Start.
4. Export the results as CSV, Excel, JSON, or XML from the Dataset tab.

Run it programmatically through the Apify API (run-sync-get-dataset-items) or the ApifyClient for JavaScript and Python.

Use with AI agents (MCP)

Give an AI agent live access to Replicate through the Model Context Protocol. Add the Actor to Claude, Cursor, or any MCP client:

claude mcp add --transport http apify "https://mcp.apify.com?tools=parseforge/replicate-models-scraper"

Then prompt it in plain language to run the scraper and read back the results.

Troubleshooting

Why am I getting no results?

Check that the collection slug is spelled correctly and matches an existing Replicate collection. If left empty, the Explore page will be scraped. Also ensure 'Maximum models' is set to at least 1.

Why does the run stop before reaching my maximum models?

The Actor stops when it has collected all available models on the page. If the collection or Explore feed has fewer models than your maximum, the run ends early.

Some model fields are empty in my dataset.

Not every model listing includes all metadata. If a field like hardware or description is missing on Replicate, it will appear empty in your results.

The scraper is running slowly.

Large runs with a high maximum model count take longer because the Actor scrolls through many pages. Reduce the maximum or run it less frequently.

I get an error about an invalid collection slug.

Visit replicate.com/collections to see the available collections. Use the exact slug from the URL, for example 'text-to-video' not 'text to video'.

FAQ

Question	Answer
Do I need a Replicate API key to use this scraper?	No. This Actor reads the public model listing pages on replicate.com, so no API key or account is required.
What is a collection slug?	A collection slug is the last part of a Replicate collection URL, for example 'text-to-image' in replicate.com/collections/text-to-image. Use it to scrape only models from that category.
Can I scrape all models on Replicate?	You can scrape up to 1,000,000 models per run from the Explore feed or a specific collection. The Explore feed includes featured, official, and trending models.
What data fields does the scraper return?	Each row includes the model name, description, run count, owner, hardware information, and other metadata visible on the listing page.
How do I scrape only the newest models?	Set a low 'Maximum models' value and run the Actor frequently. The Explore feed and collection pages list models with newer and trending ones first.
Can I filter models by run count or owner?	The Actor collects all models from the chosen source. To filter by run count or owner, export the dataset and filter it in a spreadsheet or database.
Is this an official Replicate integration?	No, this is an unofficial scraper built by ParseForge that reads public web pages. It is not affiliated with Replicate, Inc.
What export formats are supported?	You can export your dataset in CSV, JSON, Excel, or XML from the Apify platform.
How often can I run this Actor?	You can schedule it to run as often as you need, from once to every few minutes, using Apify's built-in scheduler.
Does this scraper handle pagination automatically?	Yes. The Actor scrolls through the model feed and collects entries until it reaches your set maximum or the end of the list.

Related actors

Browse the full ParseForge collection for more scrapers.

🆘 Need help? Email parseforge@protonmail.com with your run ID, your input, and what you expected.

⚠️ Disclaimer. This Actor is unofficial and is not affiliated with, endorsed by, or sponsored by Replicate, Inc. It collects only publicly available data. You are responsible for using the collected data in compliance with the source's terms of service and applicable data-protection laws, including GDPR, CCPA, and PIPL. Do not use it to collect personal data unlawfully.