Fal.ai Models Scraper

Extract the complete catalog of AI models from fal.ai — including pricing, categories, model labs, API endpoints, license types, and more. Monitor 1,300+ models across text-to-image, image-to-video, text-to-speech, and every other inference category on the platform.

No browser required. Uses the official fal.ai JSON API with fast pagination and optional filtering by category or keyword.

What does it do?

This actor fetches every model listed on fal.ai using their paginated REST API. For each model it extracts:

• 🆔 Model path — unique identifier (e.g. fal-ai/flux/schnell)
• 📝 Title and short description
• 📂 Category (text-to-image, image-to-video, speech-to-text, etc.)
• 🏷️ Tags (e.g. realism, typography, open-source)
• 🔗 API URL and landing page URL
• 💰 Pricing info (cost per image/run as published by fal.ai)
• 🏢 Model lab / provider (Google, Stability AI, Black Forest Labs, etc.)
• 🔑 License type (commercial, open-source, etc.)
• 🖥️ Hosting type (proxy, native)
• 📅 Published date and deprecation status
• 🖼️ Thumbnail URL and GitHub URL (when available)

Pagination is handled automatically — all 34+ pages are fetched in sequence. You can cap the result count or filter by category or keyword.

Who is it for?

AI developers & engineers — Building on fal.ai or comparing it against Replicate, RunPod, or Hugging Face Inference? Scrape the full catalog programmatically to search, compare, and integrate without clicking through the web UI.

Competitor analysts & product managers — Track which model labs (Google, Stability AI, Black Forest Labs) have the most models on fal.ai. Monitor pricing changes as new models launch. Export the full catalog to a spreadsheet or BI dashboard.

ML researchers — Catalog model availability, license types, and API endpoints across inference providers. Correlate fal.ai model launches with paper releases or GitHub activity.

Automation builders — Trigger workflows whenever new models are added in a specific category (e.g. text-to-video) using Apify schedules + Make/Zapier/n8n integrations.

Data aggregators — Combine fal.ai model data with Replicate, Hugging Face, and Civitai data for a unified AI model catalog.

Why use this actor?

• ✅ Zero browser overhead — uses fal.ai's official JSON API, 10× faster than HTML scraping
• ✅ Full catalog — fetches all 1,300+ models automatically across all pages
• ✅ Structured output — clean JSON/CSV with 18 fields per model, ready for analysis
• ✅ Flexible filtering — filter by category (text-to-image, image-to-video, etc.) or keyword before export
• ✅ No proxy required — the fal.ai API is publicly accessible, no residential proxy cost
• ✅ No proxy required — the fal.ai API is publicly accessible, so there are no residential proxy costs

Data extracted

Field	Description	Example
modelPath	Unique model identifier	fal-ai/flux/schnell
title	Display name	FLUX.1 [schnell]
category	Inference category	text-to-image
tags	Associated tags	["open-source", "speed"]
shortDescription	One-line description	Fast 4-step text-to-image model
modelUrl	API call endpoint	https://fal.run/fal-ai/flux/schnell
landingPageUrl	Model landing page	https://fal.ai/models/flux-schnell
thumbnailUrl	Preview image URL	https://v3b.fal.media/...
githubUrl	GitHub repo (if available)	https://github.com/black-forest-labs/flux
licenseType	License	open-source
status	Availability	public
deprecated	Is model deprecated?	false
publishedAt	ISO 8601 publish date	2024-08-01T12:00:00.000Z
modelLab	Creator / provider	Black Forest Labs
modelFamily	Model family	FLUX.1
groupLabel	Display group	Image Generation
hostingType	How it's hosted	proxy
kind	Model kind	inference
pricingInfo	Pricing text	$0.003 per image...

How much does it cost to scrape fal.ai models?

This is a pure HTTP actor — no browser, no proxies. Costs are very low.

Run type	Models scraped	FREE tier cost	DIAMOND tier cost
Quick sample	50 models	~$0.063	~$0.019
Full catalog	1,300+ models	~$1.50	~$0.37
Category only	~50–200 models	~$0.063–$0.24	~$0.019–$0.061

Pricing uses a pay-per-result model:

• 🚀 Start fee: $0.005 per run (one-time)
• 📄 Per model: tiered pricing from $0.00115 (FREE tier) to $0.00028 (DIAMOND tier for high-volume users)

Higher-volume Apify subscription tiers pay significantly less per result. DIAMOND users (highest volume) pay ~4× less per model than FREE users.

How to scrape fal.ai models — step-by-step

1. Open the actor in Apify Store
2. Click Try for free
3. Configure your inputs (or keep the defaults):
   • Set Max models to the number you want (default: 200; use 0 for all)
   • Optionally filter by Category (e.g. text-to-image)
   • Optionally set a Keyword filter (e.g. stable diffusion)
4. Click Start and wait ~5–30 seconds
5. Download results as JSON, CSV, or Excel

Input parameters

Parameter	Type	Description	Default
maxResults	Integer	Maximum models to extract	200
category	String	Filter by category (e.g. text-to-image)	All
searchQuery	String	Keyword filter (matches title, description, tags, lab)	None
maxRequestRetries	Integer	HTTP retry attempts	3

Output example

{
  "modelPath": "fal-ai/flux/schnell",
  "title": "FLUX.1 [schnell]",
  "category": "text-to-image",
  "tags": ["open-source", "speed"],
  "shortDescription": "FLUX.1 [schnell] is a 12B parameter rectified flow transformer for rapid text-to-image generation.",
  "modelUrl": "https://fal.run/fal-ai/flux/schnell",
  "landingPageUrl": "https://fal.ai/models/flux-schnell",
  "thumbnailUrl": "https://v3b.fal.media/files/flux/preview.jpg",
  "githubUrl": "https://github.com/black-forest-labs/flux",
  "licenseType": "open-source",
  "status": "public",
  "deprecated": false,
  "publishedAt": "2024-08-01T12:00:00.000Z",
  "modelLab": "Black Forest Labs",
  "modelFamily": "FLUX.1",
  "groupLabel": "Image Generation",
  "hostingType": "proxy",
  "kind": "inference",
  "pricingInfo": "Your request will cost $0.003 per image..."
}

Tips for best results

• 🔢 Get the full catalog: leave category empty and set maxResults to 2000 (the API has ~1,300 models as of 2026)
• 📂 Track a category: set category = text-to-video and run on a weekly schedule to monitor new video models
• 🔍 Find models by provider: set searchQuery = Google to get only Google-created models
• 📉 Find deprecated models: download the full catalog and filter deprecated = true in your spreadsheet
• 🔄 Automate monitoring: combine with Apify Schedules + Make/Zapier to get Slack/email alerts when new models appear

Integrations

Apify + Make (Zapier)

Use this actor as a scheduled trigger in Make to monitor fal.ai for new model launches:

1. Create a Make scenario with the Apify → Watch Dataset Items trigger
2. Schedule the Fal.ai Models Scraper to run weekly
3. Add a filter: publishedAt > last_run_date
4. Send new model notifications to Slack, email, or Notion

Google Sheets

Export the full model catalog to Google Sheets for analysis:

1. Run the actor and download as CSV
2. Import to Google Sheets: File → Import → Upload
3. Use =FILTER(A:A, J:J="text-to-image") to slice by category
4. Add conditional formatting to highlight deprecated models (K:K = TRUE)

Airtable / Notion

Use the Apify → Airtable or Apify → Notion integrations in Make or Zapier to sync model data directly into a database view for team collaboration.

Python pipeline

import requests

run = requests.post(
    "https://api.apify.com/v2/acts/automation-lab~fal-ai-models-scraper/runs",
    headers={"Authorization": "Bearer YOUR_API_TOKEN"},
    json={"maxResults": 500, "category": "text-to-image"}
).json()
run_id = run["data"]["id"]

# Poll until done, then fetch data
items = requests.get(
    f"https://api.apify.com/v2/actor-runs/{run_id}/dataset/items",
    headers={"Authorization": "Bearer YOUR_API_TOKEN"}
).json()

API usage

Node.js

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: 'YOUR_API_TOKEN' });

const run = await client.actor('automation-lab/fal-ai-models-scraper').call({
    maxResults: 500,
    category: 'text-to-image',
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(`Extracted ${items.length} models`);

Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_API_TOKEN")

run = client.actor("automation-lab/fal-ai-models-scraper").call(run_input={
    "maxResults": 500,
    "category": "text-to-image",
})

for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["title"], item["modelLab"])

cURL

curl -s -X POST \
  "https://api.apify.com/v2/acts/automation-lab~fal-ai-models-scraper/runs" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"maxResults": 100, "category": "text-to-image"}' \
  | jq '.data.id'

Use with Claude (MCP)

You can use this actor directly in Claude Desktop, Claude Code, or Cursor via the Apify MCP server:

Claude Code / terminal:

$claude mcp add --transport http apify "https://mcp.apify.com?tools=automation-lab/fal-ai-models-scraper"

Claude Desktop / Cursor / VS Code — add to your MCP config:

{
  "mcpServers": {
    "apify": {
      "type": "http",
      "url": "https://mcp.apify.com?tools=automation-lab/fal-ai-models-scraper",
      "headers": { "Authorization": "Bearer YOUR_APIFY_TOKEN" }
    }
  }
}

Example prompts for Claude:

• "Scrape all text-to-image models from fal.ai and show me the ones from Black Forest Labs"
• "Get the full fal.ai model catalog and find all models with open-source licenses"
• "Fetch fal.ai models filtered by category image-to-video and export a comparison table"

Legality and terms of service

This actor uses fal.ai's publicly available REST API (/api/models) to fetch model metadata — the same data shown on the fal.ai models page. No login is required and no personal data is collected.

fal.ai's public API is intended for developers to programmatically access model information. Always review fal.ai's Terms of Service and their robots.txt before automated use. This actor does not scrape user data, authentication tokens, or any non-public endpoints.

Use of Apify's platform is subject to Apify Terms of Service. Always use data responsibly and in compliance with applicable laws.

FAQ

How many models does fal.ai have?

As of early 2026, fal.ai hosts over 1,300 models across 10+ inference categories. The catalog grows regularly as new models are added by labs like Google, Stability AI, Black Forest Labs, and others.

Does this require proxies or login?

No. The fal.ai /api/models endpoint is a public JSON API that requires no authentication or proxies. The actor makes plain HTTP requests.

Can I filter to only non-deprecated models?

The actor returns all non-private, non-removed models. After downloading, filter on deprecated = false in your spreadsheet or code to exclude deprecated models.

Can I get all models in a specific category?

Yes — set the category input to the desired category name (e.g. text-to-image, image-to-video). The supported categories are listed in the input dropdown.

I got fewer results than expected. Why?

If you set maxResults lower than the total catalog size, the actor stops early. Set maxResults to 2000 to ensure you get the full catalog. Models marked as removed or private are excluded from output.

The run finished instantly with 0 results. What happened?

If you set a category that does not exist on fal.ai (e.g. a typo), all models will be filtered out. Check the category input matches one of the available values in the dropdown.

Does this actor stay up to date?

The actor fetches live data from fal.ai on every run — there is no cached dataset. Schedule it weekly or monthly to keep your model catalog current.

Related scrapers

Looking for data from other AI model platforms? Check out our related actors:

• Replicate Models Scraper — models and pricing from Replicate
• Groq Models Scraper — LLM models available on Groq
• Cloudflare Workers AI Scraper — models hosted on Cloudflare Workers AI
• Hugging Face Models Scraper — open-source models on Hugging Face Hub