Features
OpenModeration covers the whole pipeline, from receiving a comment to acting on it.
Single Integration
One API endpoint, one response format. When we plug in a better AI model, nothing changes in your code.
Our own moderation AI
Submit content once and get back a clear pass/fail verdict and category scores, set up on your rules and your use cases.
Pre-built content categories
Ready-made categories covering hate, harassment, sexual content, violence, self-harm, spam, misinformation, PII, and more.
Cross-checked by other AIs
On delicate cases, other AI models arbitrate the decision before it is applied. Fewer mistakes, fewer wrongly hidden comments.
Smarter Moderation Decisions
Decisions take into account the conversation, the history of the author, the channel and your own policies.
Bulk Content Review
Submit up to 10,000 items at once for async processing. Results are pushed to your system when they are ready.
Custom Moderation Policies
Define your own categories. Provide examples and let the AI classify content your way.
Moderation Dashboard
A web interface to run your day-to-day moderation.
Access Management
Create and manage access keys. Set usage limits per key.
Team & Role Management
Give your colleagues access, assign roles and set permissions without going through your IT department.
Report Management
One queue to process user reports and track their status, so no flagged comment gets forgotten.
Activity History
Every moderation request is logged with its verdict, response time and category scores.
Service Health Monitoring
See the real-time status of the service: uptime, response times, and error rates.
Performance Reports
Follow moderation volume and flag rates and watch how costs evolve.
Connector Management
Connect your social platform accounts with your own credentials.
Rule Configuration
Define forbidden words, set category thresholds and write your rules in plain language. These three tools need no code.
Automated Moderation
You define your moderation rules with three tools: forbidden words, category thresholds and your own rules written in plain language.
Context-Aware AI Rules
Combine AI confidence scores with your own rules instead of relying on keyword lists alone.
Auto-Actions
When a threshold is crossed, the Action Engine can hide, delete, flag, notify or mark as spam.
Human-in-the-Loop
Borderline cases are handled by our team. You can step in and make the final call whenever you want.
Real-Time View
See every decision as it happens, with its written reason, and approve or reject any item yourself.
Webhook Notifications
Get real-time alerts for flagged content, finished batches, errors and more.
Full Audit Trail
Every action is logged with the rule that triggered it, the action taken and the outcome.
Platform Integrations
Connect directly to social platforms and your blog to automatically ingest and moderate content.
Receive comments and posts. Automatically hide or remove content that breaks your rules.
Moderate Instagram comments and messages. Keep your profile safe from toxic or spammy content.
YouTube
Ingest video comments. Mark as spam or hold for review by our team.
WordPress
Moderate comments on your blog. Approve, hold for review, or mark as spam automatically.
Use Your Own Platform Access
Connect with your own platform accounts. Access tokens are never stored in plaintext.
Generic Webhook Receiver
Ingest content from any platform that supports outgoing webhooks.
Generic Webhook Callback
Send moderation results to your own webhook for custom integrations.
One format, whatever the model behind it
When we plug in a better or cheaper AI model, results keep the same format, so you only integrate once.
The same format every time, with no extra code to write when the AI improves.
Secure access keys with scoped permissions and usage limits per key.
Interactive API docs are generated automatically, and you can try requests from your browser.
# One call, standardized scores
curl -X POST https://api.openmoderation.com/v1/moderation \
-H "Authorization: Bearer om_key_..." \
-H "Content-Type: application/json" \
-d '{
"input": "Your text to moderate here",
"language": "en"
}'
# Response (same format, every time)
{
"id": "modr_01JXK...",
"results": [{
"category": "hate",
"score": 0.02,
"flagged": false
}],
"flagged": false,
"usage": {
"latency_ms": 243
}
}Protect your brand in a few minutes.
Connect your site or your pages and tell us your rules and use cases. We set up the moderation and run it for you: our team handles the doubtful cases and reports, and you see every decision and its reason in real time.
No credit card required · DSA-compliant · GDPR-compliant
Questions? contact@openmoderation.com