Back to Moderation page

Revolutionize content safety with AI-powered image moderation

Give your users an enhanced and filtered visual experience.

Try now
Image

Ready in 3 simple steps

  • Try our pre-trained moderation model to 
    see how it works out of the box
  • Tune the model with your data and add 
    your own classes for even better results
  • Access the model via API — no need to 
    invest time and money in infrastructure

We detect content across multiple model classes

  • Derogatory language
    Disrespectful, hurtful, or hateful language targeted towards individuals or groups of people
  • Threats of violence
    Explicit, implied, or suggested acts of physical harm, aggression, or intimidation
  • Obscenity, profanity
    Vulgar language or offensive expressions that are deemed inappropriate or indecent
  • Promotions
    Advertising, endorsing, or publicizing products, services, events, or ideas in an online environment
  • Gibberish
    Nonsensical, random, or unintelligible text, which adds no value or substance to the online conversation
  • Phone numbers
    Phone numbers inserted in message texts, including international formats
  • Personal data
    Sensitive or identifiable information about individuals, such as full names, addresses, phone numbers, etc
  • Spam
    Unsolicited, irrelevant, or repetitive texts with the intent to promote, advertise, or manipulate
  • Sexual themes
    Explicit, suggestive, or inappropriate sexual material or themes, which may be discomforting or offensive
  • Politics
    Debating or sharing information related to political figures, parties, policies, and government affairs

Languages

English

Spanish

German

French

Italian

Russian

Turkish

Arabic

Hindi

How our 
platform works

Try now
  • Image
    Step 1: Choose the types of content you want to hide
    Pick your categories — threats, spam, or everything at once
  • Image
    Step 2: Fine-tune the moderation rules
    Support for a softer or stricter policy on certain content
  • Image
    Step 3: Connect via API
    Clear and fast integration
  • Image
    Step 4: Results and improvements
    Our service adapts to your data with human-in-the-loop optimization
Request a demo

Pricing

  • TRIAL
    $0
    • Get free, rate-limited usage for learning and prototyping:
    • AI moderation only
    • 100 texts per day limit
    • 6 basic classes, English only
    Start now
  • STARTUP
    $1199
    • Power your business with next-level moderation service:
    • AI moderation with human-in-the-loop quality control
    • 300k texts per month  included
    • 10 classes & 9 languages
    • E-mail support with 24-h SLA
    Contact us
  • ENTERPRISE
    Custom
    • Moderation perfectly customized for your business:
    • AI and human moderation
    • No limits
    • Custom classes & 50+ languages
    • Iterative model retraining & dedicated support
    Contact us
Additional cost for resources usage
Storage per GB/month$0.030
GPU per hour (NVIDIA V100)$10.00

Control the moderation process

View stats in the online dashboard
  • Analytics — get visualized statistics by category
  • Tune settings for accuracy — turn on manual moderation for ambiguous images and borderline cases
  • Convenient interface — available for web and mobile
Image
Image
See what's happening
  • Real-time — view all comments in your social stream and moderation results
  • Granular view — see which comments are blocked and why
  • Sort by category — group blocked comments by type (Racism, LGBTQ+ Slurs, Severe Swearing, etc)
Customize moderation rules 
in the visual interface
  • Scope and rules — adjust the confidence intervals for sending content for human moderation
  • Team management (user roles) — assign access rights and roles to users
  • Data retention policy — control whether to delete the data or store it for retraining the model
Image

Fast, no-code and human-assisted AI solutions

Start in minutes and get immediate results
Request a demo
Fractal