LLM-powered content moderation

Moderate user content in a single API call.

ModerateHQ reads your text and returns a clear verdict: approve, flag, or block, with categories and a reason. Ship trust & safety without building a model.

Get started free See pricing

Free trial · No credit card to start · 1 credit = 1 request

By signing up you agree to our Terms and Privacy Policy.

moderate.shapproved

$ curl https://api.moderatehq.com/v1/moderate/text \ -H "x-api-key: $MODERATEHQ_KEY" \ -H "Content-Type: application/json" \ -d '{ "content": "I'm going to kill you", "policy": "gaming", "context": { "platform": "my-shooter", "channelType": "ranked match chat", "conversationHistory": [ "gg that round was so close", "haha you clutched it, rematch?" ] } }' { "verdict": "approved", "confidence": "high", "categories": [], "explanation": "In a ranked match this reads as trash talk about an in-game kill, not a real-world threat.", "policyApplied": "gaming" }

Drop the match context and the same line is flagged for human review.

Trash talk, read in context · gaming policy

Everything you need to keep content clean

A focused moderation API with the guardrails and visibility a production trust & safety workflow needs.

One call, one verdict

POST your text and get back approved, flagged, or blocked: a categorical decision you can act on directly.

Verdicts in seconds

Most verdicts return in a second or two. If one can't be produced quickly, you get a fast, clear error instead of a hanging request.

Pay per request

Simple credit model: 1 credit = 1 request. Top up when you need to, no seats, no monthly minimum. Start on a free trial.

Context-aware moderation

Send the surrounding context: platform, channel type, and recent conversation. The same words are then judged for where they actually appear, not in isolation.

Policies that fit you

Pick a built-in policy that matches your platform, from general communities to strict or gaming, and tune how strict it is. The rules fit your product, not the other way around.

Full visibility

A dashboard for keys, usage, credit history, and a searchable moderation log. See exactly what was decided and why.

Simple, usage-based pricing

Buy credits, spend one per request. No subscriptions, no seats. Top up whenever you need more.

Paid plans are coming soon. Every account starts on a free trial you can use today. Credit purchases switch on here once they are ready.

Start free trial

Starter

For small projects and getting started.

$10one-off

2,500 credits

$4.00 per 1,000 requests

Credits stay valid while your account is active

Coming soon

Pro

For apps with steady moderation volume.

$50one-off

15,000 credits

$3.33 per 1,000 requests

Credits stay valid while your account is active

Coming soon

Scale

Best per-request rate for high throughput.

$200one-off

70,000 credits

$2.86 per 1,000 requests

Credits stay valid while your account is active

Coming soon

All paid plans run at up to 5 requests per second. Just exploring? Every new account starts on a free trial: 10 requests per day at up to 2 requests per second, no card required.

Frequently asked questions

How does pricing work?

You buy credits in packs. One credit equals one moderation request, charged only on a successful verdict. Credits stay on your balance while your account is active, and only expire after a long stretch of inactivity (at least 12 months with no sign-ins or API calls). We email you well before that happens.

Is there a free trial?

Yes. Sign up and you get a trial account with a daily request allowance and an API key created for you on first login. No card required to start.

What can it moderate?

Text today: approve, flag, or block against your policy, with categories and a short explanation on every verdict. Image and audio moderation are on the roadmap.

What does a 'flagged' verdict mean?

Borderline content the model isn't confident is safe or unsafe. Rather than auto-blocking, you route flagged items to human review. The verdict is the source of truth, not a raw confidence number.

How do I integrate it?

Send an HTTP POST with your API key and the content to evaluate. You get structured JSON back. That's the whole integration.

Do you store the content I send?

By default, content and explanations are kept for 30 days so you can review them in your moderation log, then deleted. You can also turn content storage off entirely in Settings: nothing you send is stored, and your log keeps verdicts and statistics only.

Start moderating in minutes

Create an account, grab your API key, and send your first request. The free trial is waiting.

Get started free