Civitai Models & Creators Scraper avatar

Civitai Models & Creators Scraper

Pricing

from $3.00 / 1,000 results

Go to Apify Store
Civitai Models & Creators Scraper

Civitai Models & Creators Scraper

Scrape Civitai, the largest AI art model sharing platform. Search AI models (Checkpoints, LoRAs, ControlNets), browse creators, collect generated images with full metadata, and explore tags. Supports cursor-based pagination for large result sets

Pricing

from $3.00 / 1,000 results

Rating

5.0

(7)

Developer

Crawler Bros

Crawler Bros

Maintained by Community

Actor stats

7

Bookmarked

2

Total users

1

Monthly active users

2 days ago

Last modified

Share

Search and extract AI models, creators, generated images, and tags from Civitai — the world's largest open AI art model sharing platform.

What does this actor do?

Civitai hosts hundreds of thousands of community-trained AI models including Stable Diffusion Checkpoints, LoRAs, ControlNets, Embeddings, and more. This actor lets you:

  • Search AI models by keyword, type, rating, and time period
  • Browse creators and their published model counts
  • Collect generated images with full generation metadata (prompt, seed, sampler, CFG scale)
  • List tags used to categorize models on the platform
  • Fetch specific models by ID or Civitai URL
  • Get all models by a creator username

Data is fetched directly from the official Civitai REST API using cursor-based pagination to reliably retrieve large result sets.

Input

FieldTypeDescriptionDefault
modeSelectWhat to fetch: searchModels, searchCreators, searchImages, getTags, getModel, byUsersearchModels
searchQueryTextFree-text search query
modelIdsString ListModel IDs for direct lookup (mode=getModel)
startUrlsString ListCivitai model URLs for direct lookup (mode=getModel)
usernameTextCreator username for mode=byUser or image filter
modelTypeSelectFilter by type: Checkpoint, LORA, ControlNet, VAE, etc.
sortBySelectSort order: Highest Rated, Most Downloaded, NewestNewest
periodSelectTime period: AllTime, Year, Month, Week, DayAllTime
nsfwAllowedBooleanInclude NSFW content in resultsfalse
minRatingNumberMinimum average rating (0–5)
maxItemsIntegerMaximum number of records to emit (1–5000)50
apiTokenSecretOptional Civitai API token for higher rate limits

Output

Model record

FieldTypeDescription
modelIdIntegerUnique Civitai model ID
nameStringModel name
typeStringModel type (Checkpoint, LORA, etc.)
nsfwBooleanWhether the model is marked NSFW
ratingFloatAverage community rating (0–5)
ratingCountIntegerNumber of ratings
downloadCountIntegerTotal number of downloads
favoriteCountIntegerTotal favorites
commentCountIntegerTotal comments
tagsArrayList of tag strings
creatorObjectCreator username and image URL
versionsArrayModel versions with download URLs and preview images
urlStringDirect Civitai URL for the model
recordTypeStringAlways "model"
scrapedAtStringISO 8601 timestamp of extraction

Creator record

FieldTypeDescription
usernameStringCivitai username
avatarUrlStringProfile image URL
modelCountIntegerNumber of published models
urlStringProfile page URL
recordTypeStringAlways "creator"
scrapedAtStringISO 8601 timestamp

Image record

FieldTypeDescription
imageIdIntegerUnique image ID
urlStringDirect image URL
nsfwStringNSFW level: None, Soft, Mature, X
widthIntegerImage width in pixels
heightIntegerImage height in pixels
metaObjectGeneration metadata: prompt, seed, steps, sampler, cfgScale
statsObjectReaction counts: likeCount, heartCount, etc.
usernameStringCreator username
recordTypeStringAlways "image"
scrapedAtStringISO 8601 timestamp

Example output record

{
"modelId": 12345,
"name": "Ultra Realistic Portrait v2",
"type": "Checkpoint",
"nsfw": false,
"rating": 4.87,
"ratingCount": 512,
"downloadCount": 198000,
"favoriteCount": 5400,
"commentCount": 220,
"tags": ["portrait", "realistic", "photography"],
"creator": {
"username": "portrait_master",
"image": "https://image.civitai.com/avatar.jpg"
},
"versions": [
{
"id": 67890,
"name": "v2.0",
"baseModel": "SD 1.5",
"downloadUrl": "https://civitai.com/api/download/models/67890",
"images": ["https://image.civitai.com/img1.jpg"]
}
],
"url": "https://civitai.com/models/12345",
"recordType": "model",
"scrapedAt": "2026-05-17T12:00:00+00:00"
}

FAQ

Do I need an API token? No. The Civitai API is publicly accessible without authentication. An optional API token from civitai.com/user/account increases your rate limits and may be required to access NSFW content above the default threshold.

How many results can I get? Up to 5,000 records per run using maxItems. The actor uses cursor-based pagination to retrieve all available results beyond the first page.

How fresh is the data? Data is fetched live from the Civitai API at the time of the actor run. Model download counts and ratings reflect the state at that moment.

Can I fetch a specific model I know the URL for? Yes. Use mode=getModel and provide either the Civitai model URL (e.g. https://civitai.com/models/12345) or just the numeric ID in modelIds.

Why are some NSFW models not appearing? By default nsfwAllowed is false. Enable it to include NSFW content. For full NSFW access, provide an API token from a Civitai account with NSFW permissions enabled.

What model types are supported? All Civitai model types: Checkpoint, TextualInversion (Embedding), Hypernetwork, AestheticGradient, LORA, LoCon, ControlNet, Upscaler, MotionModule, VAE, Poses, Wildcards, Workflows, and Other.

Can I get all models by a specific creator? Yes. Use mode=byUser and set username to the creator's Civitai username.

Limitations

  • The Civitai API enforces rate limits on unauthenticated requests. Providing an API token increases the allowed request rate.
  • Model file weights and binary data are not downloaded — only metadata and preview image URLs are extracted.
  • Some NSFW content requires both nsfwAllowed=true and a valid API token from an account with appropriate settings.
  • The versions[].downloadUrl links point to the Civitai CDN. Downloading model files requires a Civitai account in some cases.