Predict

AI predictions for your images, video and text

Sign up now for free

Predict uses AI to tell you what is in your images, videos or text

Images

Capture data about the physical world through images. Our APIs support the most common formats (JPEG, PNG, BMP, TIFF, WEBP). Responses fire in 50-500ms intervals, depending on complexity.

Video

Process video 100x faster than a human ever could. Our APIs operate on video clips in standard formats (MP4, AVI, GIF, MOV, WMV and 3GPP). Outputs are encoded as a time series for easy analysis. 

Text

Analyze text in real-time and offer responses to a conversation without missing a beat. Categorize, sort and search documents at scale. We help you unlock the meaning behind your written content.

 

Customers use Predict for:

Moderation_gr

Moderation

visual_search_gr

Visual Search

product_discovery_gr

Product Discovery

automated_image_video_tag_gr

Automated Image & Video Tagging

OCR_gr

Optical Character Recognition

face_detection_gr

Face Detection

face_verification_gr

Face Verification

sentiment_analysis_gr

Sentiment Analysis

audience_insights

Audience Insights

logo_detection_gr

Logo Detection

brand_safety_gr

Brand Safety

document_organization_gr

Document Classification

quality_control_gr

Quality Control

employee_safety_gr

Employee Safety

security_monitoring_gr

Security Monitoring 

aerial_surveillance-1

Aerial Surveillance

Unlock AI inference with Models and Concepts

Models

Models are functions that take inputs and produce outputs. Clarifai models can be combined into distinct workflows. Clarifai offers models in two flavors:

  • Machine learning models that learn patterns and produce different outputs based on these patterns
  • Fixed function models that produce outputs based on a predetermined set of parameters
illustration model training

Concepts

When you call the Predict API, you are asking it to predict the likelihood that your data contains a given concept. Each model can recognize a finite number of concepts. When you create a custom model, you are teaching the model to recognize new concepts.

Gain insights on your data with multiple prediction types

 

Classification

Categorize large quantities of unstructured image, video and textual data faster than has ever been possible before. Classify objects, faces, people, sentiment, and more with state of the art AI. Leverage a suite of pre-trained models or train a custom classification model to suit your needs.

furniture-classification-zoom

Base Classification Models

We’ve trained our models on millions of images so you don’t have to.

 

Browse our model gallery >

General

Travel

Food

Wedding

Moderation

Apparel

Real estate

Furniture

Textures 

Art styles

 

Vehicles

Dog breeds

Hair style

Roof type

Detection

Locate the people, places and things within your images, videos, and live feeds. Detection takes our Classification technology a step further by identifying the exact placement of concepts within your image or video.

Face detection on images and video

Base Detection Models

We’ve trained our models on millions of images so you don’t have to.

 

Browse our model gallery >

General 

Apparel

Animal

Text

Logo

People

Face

Vehicle

Furniture

Rooftop

Vents

Rooftop HVAC

Customized Classification

Solve your unique business problem with custom models. Simply provide some example training data and label the things or people you want to recognize. You’ll have an AI model that can predict exactly what you want in seconds.

 

Learn more about custom training >

See pricing >

jump_for_joy-1

Face Recognition

Recognize who is in your photos, videos and live streams.

Detect specific people to verify their identity. Identify faces with as few as one training sample with our advanced pre-trained facial recognition models. Get started in minutes with a pre-trained model that can recognize famous celebrities.

Understand your customers by identifying other facial attributes such as age, gender and multicultural appearance.

See pricing >

Face detection and face recognition on images and video

Embedding Vectors

Leverage our deep understanding of images, videos and text in your machine learning tasks with embedding vectors. Train on top of these vectors, or compare them with your distance metric of choice to find similarity between data inputs. We strive to empower your teams to solve their most challenging business problems, and unlike other AI platforms, we let you export these features.

Embedding vectors and concept segmentation image

Fixed Function Models

Build complex use cases in our API with handy fixed function models. Examples include random sampling to filter down data stream, image cropping to select a region from an image, and concept thresholding to pass predictions through thresholds.

Pre-Built Fixed Function Models

We have a number of fixed function models to get you started. Email product-feedback@clarifai.com to suggest a new function. 

Fixed function models

Multilingual Classification

Deliver a consistent user experience for all your customers regardless of their native language. Powered by Clarifai’s extensive knowledge graph, multilingual classification can "see" in 23 different languages to support your global enterprise.

flag_detection

23 Supported Languages

Arabic

Bengali 

Danish 

German 

English 

Dutch

Norwegian 

Punjabi 

Polish 

Portuguese

Hindi

Hungarian

Italian

Japanese

Korean

Russian

Swedish

Turkish

Chinese Simplified

Chinese Traditional

Spanish 

Finnish 

French

Clarifai partners with organizations of all sizes to deploy cutting-edge artificial intelligence

World leading AI

  • Forrester identified Clarifai as an industry leader
  • Research scientists named on over 1250 publications drawing over 140K citations
  • Dozens of pre-trained models deliver incredible accuracy from day one
  • Knowledge graph linking for concepts in multiple languages
  • GDPR and Privacy Shield compliant

Up & running in minutes

  • Get started with 5,000 free operations each month, no credit card required
  • Use the fastest  UX to add your inputs and train your models
  • Take advantage of up-to-date developer documentation and customer-first support
  • Program with API clients in six languages
  • Leverage RESTful and gRPC APIs for ease of use and high performance

Scaleable for your business

  • Cloud, on-premise, or air-gapped: We deploy to fit your needs
  • Fine-grained permissions let your organization collaborate
  • Models get smarter and more accurate as data flows into our API
  • Fast responses enable real-time applications
  • Output auto-scales to any incoming bandwidth

Built for developers from the ground up

Clarifai was founded by developers and designed to help developers start building immediately and quickly master AI

curl -X POST
-H
'Authorization: Key YOUR_API_KEY'
-H
"Content-Type: application/json"
-d
'
{
"inputs": [
{
"data": {
"image": {
"url": "https://samples.clarifai.com/metro-north.jpg"
}
}
}
]
}'

https://api.clarifai.com/v2/models/e0be3b9d6a454f0493ac3a30784001ff/outputs

 

app.models.initModel({id: Clarifai.GENERAL_MODEL, version: "aa7f35c01e0642fda5cf400f543e7c40"})
.then(generalModel => {
return generalModel.predict("the-image-url");
})
.then(response => {
var concepts = response['outputs'][0]['data']['concepts']
})
from clarifai.rest import ClarifaiApp
from clarifai.rest import Image as ClImage
app = ClarifaiApp(api_key='your_api_key')
model = app.public_models.general_model
model.model_version = 'aa7f35c01e0642fda5cf400f543e7c40'
response = model.predict(Image(url="https://samples.clarifai.com/metro-north.jpg"))
ConceptModel model = client.getDefaultModels().generalModel();
ModelVersion modelVersion = model.getVersionByID("the-version").executeSync().get();

ClarifaiResponse<List<ClarifaiOutput<Prediction>>> response = client.predict(model.id())
.withInputs(ClarifaiInput.forImage("some-URL"))
.withVersion("aa7f35c01e0642fda5cf400f543e7c40")
.executeSync();
ClarifaiImage *image = [[ClarifaiImage alloc] initWithURL:@"https://samples.clarifai.com/metro-north.jpg"];
[_app getModelByName:@"general-v1.3" completion:^(ClarifaiModel *model, NSError *error) {
[model predictOnImages:@[image]
completion:^(NSArray<ClarifaiSearchResult *> *outputs, NSError *error) {
NSLog(@"outputs: %@", outputs);
}];
}];


Predict pricing summary

Pay as you go. Our predict pricing scales with you as your business grows.

Model
All prices per 1,000 operations

Free tier limit operations First 500,000,000 operations Next 4.5 million operations Next 45 million operations Over 50 million operations
  (all prices quoted per 1,000 operations and tiers calculated monthly)
Classification 5,000 $1.20 $1.10 $1.00 $0.90
Detection 5,000 $2.00 $1.80 $1.60 $1.50
Custom classification 5,000 $3.20 $3.10 $3.00 $2.90
Face recognition 5,000 $1.20 $1.10 $1.00 $0.90
Embedding vectors 5,000 $1.20 $1.10 $1.00 $0.90
Fixed function 5,000 $1.20 $1.10 $1.00 $0.90
  Free tier limit operations First 500,000,000 operations Next 4.5 million operations Next 45 million operations Over 50 million operations
  (all prices quoted are monthly per concept in a custom model)
Custom model hosting 10 $0.50 $0.50 $0.50 $0.50

Committed Use Discounts

Commit to a monthly volume and receive a significant discount beyond standard volume discounts.
Contact our sales team to learn more.

Talk to sales