Communities & comments
Set rules for abuse and promotional content to support constructive conversations.
AI MODERATIONS · TEXT MODERATION
Text moderation.
Your rules.
Turn content signals into clear decisions, with moderation policies built around your product.
Get 1000 free credits · No payment method required
MODERATION / 01PRODUCT EXAMPLE
INPUT
This was really helpful. Thank you!
Analyze → Apply policy
✓
DECISION
PASSAllow content
No blocking rule triggered in this policy
PolicyStandard moderation
Decision basisEnabled labels + thresholds
Clear decisions. Traceable rules.
219Fine-grained labels
PASS / BLOCKActionable decisions
API + ConsoleFrom testing to integration
USE CASES
Different content.
The right boundaries.
Manage separate policies for comments, messages and AI-generated text, all in one console.
Messaging & support
Configure thresholds for contact details, transactions and harmful language based on your context.
AI inputs & outputs
Moderate prompts and responses with separate policies before and after generation.
POLICY MANAGEMENT
Every rule.
Under your control.
Start from the default policy. Adjust labels and thresholds, test changes and publish a traceable version.
Explore the product →LabelThresholdBlock
Severe personal insults
0.85Friend requests
0.90Parenting
0.80QUICK START
Start with one request.
Create an account & key
Sign up for free credits and create an API key.
Choose a policy
Use the default policy or customize a copy.
Get a decision
Get a decision, matching labels and the policy version.
FAQ
Before you get started.
Can I start without configuring a policy?
Yes. Requests without a policy or key binding use the published platform default.
How are credits calculated?
One credit per 1,024 characters, rounded up. Failed checks do not consume credits.
Can I inspect all label scores?
Yes. Set include_scores=true to receive all 219 label scores.