REST API · Safety-critical

NSFW Moderation API — CSAM detection and 2257 compliance

Q: How do I authenticate?

Bearer token in the Authorization header. Generate keys from your dashboard — sandbox for testing with sample datasets, production with rate limits and audit log enabled. Keys are revocable, support per-key IP allowlists and per-key audit log isolation.

Q: How does CSAM detection work?

Three layers: perceptual hash (PhotoDNA) matching against NCMEC databases of known CSAM; classifier inference for novel CSAM; and perceptual hash against a continuously-updated internal database. 99.96% recall on held-out test set, <0.01% false-positive rate. Detections trigger automated NCMEC CyberTipline report scaffolding for legally-required reporting.

Q: What's the rate limit?

Shared-fleet default is 100 concurrent requests per key and 1M calls per 24h. Volume tiers lift both. Dedicated endpoints have no hard cap — limited by your GPU(s) and configured queue depth. Batch submission supports up to 1,000 items per request.

Q: How accurate is age estimation?

Age estimation returns a point estimate plus 95% confidence interval; content with estimated age under 18 or confidence interval crossing 18 is flagged for human review. We do not represent age estimation as definitive legal proof of age — it's a triage tool that routes content to your human verification pipeline.

Q: Does this help with 2257 compliance?

Yes — per-content record stubs export consent verification, performer IDs, content hashes and timestamps. We're not the 2257 custodian of record for your business (you are), but our exports are designed in consultation with adult-industry counsel to make processor and audit conversations survivable. Counsel sign-off still recommended.

Q: What happens when CSAM is detected?

Detection blocks the content from publishing, logs to an immutable audit trail, fires a webhook with the detection details, and scaffolds an NCMEC CyberTipline report for your legally-required reporting. We do not auto-submit to NCMEC — you review and submit, but our scaffolding removes 80–90% of the manual report effort.

Q: Can I self-host?

Bring-your-own-GPU deployments are available where we operate the moderation stack on your infrastructure under a managed-services contract. Pure code licensing is available for larger operators with existing trust-and-safety teams. Most customers start on the shared API and move to dedicated when volume justifies it.

Q: Can I get a dedicated endpoint?

Yes. Dedicated L40S clusters start at $2,500/mo with custom thresholds, isolated audit log and dedicated NCMEC reporting pipeline. Multi-region failover available for enterprise. Talk to compliance for a quote tailored to your volume and risk profile.

Q: How is the API priced?

$0.004 per image call, $0.001 per 1K text tokens, $0.012 per video second. Volume tiers drop 40–60%. Dedicated deployments are a flat monthly fee with included call allowance and overage at volume-tier pricing. Batch + async endpoints have no minimum and no setup fee.

A multi-modal moderation API built for adult platforms: CSAM detection (perceptual hash + classifier), age estimation, explicit-content classification, category tagging and 2257 record helpers. Single endpoint handles text, image and video. PhotoDNA + NCMEC integration, automated CyberTipline report scaffolding, and per-content audit exports your processor and counsel can actually use. From $0.004 / call.

Get API Access View documentation

4M+

Decisions daily

<150ms

Image p95 latency

99.96%

CSAM recall

$0.004

Per moderation call

TL;DR

Endpoint: POST https://api.nsfwcoders.com/v1/moderate — multi-modal input (text, image, video), returns classification, age estimate, CSAM score, 2257 record stub and action recommendation.

Models: PhotoDNA + perceptual hash matching against NCMEC databases; multi-modal classifier for age estimation, explicit content, fetish categories and consent cues.

Pricing: $0.004 / call for images, $0.012 / call for video seconds, $0.001 / 1K text tokens. Volume tiers and dedicated deployments available.

SLA: 99.96% CSAM recall on held-out test set, <150ms image p95 latency, 99.9% uptime with multi-region failover. Compliance reports exportable per-content and per-period.

Overview

Moderation that protects your business — not just your users

Adult platforms face a different moderation problem than mainstream ones: not 'is this adult?' (you want it to be) but 'is this adult content legal, consented-to, and within the categories your processors allow?'. Generic moderation APIs (OpenAI, Perspective, AWS Rekognition) treat all adult content as a violation — useless when adult is your product. Our NSFW Moderation API is built for the inverse: classify, verify and document adult content, not refuse it.

The API handles text, image and video in one endpoint. CSAM detection runs perceptual hash (PhotoDNA) + classifier against NCMEC databases; age estimation estimates performer age with confidence intervals; explicit-content classification tags fetish categories, intensity levels and consent cues; 2257 record helpers export per-content audit stubs with consent verification, performer IDs and timestamps.

Who uses it: adult platforms moderating UGC at scale, cam sites verifying performer age and consent, marketplaces preventing prohibited categories (underage, non-consensual, banned fetishes), companion apps filtering AI-generated CSAM attempts, and payment processors requiring moderation evidence before approval. If your platform has UGC and processors, this API keeps both conversations survivable.

What makes it different: built for adult platforms, not against them. The moderation output is designed in consultation with adult-industry counsel — usable in processor approval, hosting conversations and law-enforcement contexts. We run the same moderation stack on our own companion and white-label platforms, so the operational bar is real traffic, not a demo dataset.

Multi-modal: text, image and video in one endpoint

CSAM detection via PhotoDNA + perceptual hash + NCMEC database matching

Age estimation with confidence intervals and verification hooks

Explicit-content classification: category, intensity, consent cues

2257 record helpers with consent verification + performer IDs

Automated NCMEC CyberTipline report scaffolding for legal compliance

NSFW Moderation API — CSAM detection — engineered by NSFW Coders

60-day delivery

prototype to production

Product showcase

What your users actually see

A live-grade interface built on the same components we ship to production — designed for retention, monetization, and scale.

NSFW moderation API dashboard with CSAM detection and age estimation by NSFW Coders — CSAM detection, age estimation and 2257 record helpers — the moderation layer your processor actually wants to see.

Who it's for

Who calls this API

Adult UGC platforms

Moderate user uploads at scale: classify, age-verify, document and route.

Cam sites

Performer age verification, consent documentation and live-stream moderation.

Adult marketplaces

Block prohibited categories (underage, non-consensual, banned fetishes) automatically.

AI companion apps

Filter AI-generated CSAM attempts and prohibited content in user prompts.

Payment processors

Require moderation evidence before approving high-risk merchant accounts.

Adult hosting providers

Pre-publish moderation for hosted adult content and abuse-team triage.

Why NSFW Coders

Why this moderation API vs generic

/ 01

Built for adult content

Classifies and documents — doesn't refuse. Generic APIs treat all adult as a violation.

/ 02

Multi-modal in one call

Text + image + video in a single endpoint, not three separate APIs to stitch together.

/ 03

CSAM detection done right

PhotoDNA + perceptual hash + classifier + NCMEC database matching — 99.96% recall on held-out test.

/ 04

2257 record helpers

Per-content audit stubs with consent verification, performer IDs and timestamps for processor conversations.

/ 05

Counsel-designed output

Reports structured in consultation with adult-industry counsel — usable in legal and processor contexts.

/ 06

Operated at scale

The same moderation stack that protects our own companion and white-label platforms — 4M+ decisions daily.

Authority & track record

The moderation layer we built for ourselves

now offered as an API — battle-tested on real adult traffic

Multi-modal classifiers

Text, image and video moderation in one API — CSAM detection, age estimation, explicit-content classification and category tagging.

Perceptual hash + NCMEC

PhotoDNA and pHash matching against known CSAM databases; automated NCMEC CyberTipline report scaffolding for legal compliance.

2257 record helpers

Per-content record exports with consent verification, performer IDs, content hashes and timestamp logs for audit conversations.

Operated at scale

The same moderation stack that protects our own companion and white-label platforms — 4M+ moderation decisions daily.

Compliance-friendly output

Reports designed in consultation with adult-industry counsel — usable in processor, hosting and law-enforcement conversations.

What's included

API features what ships in the endpoint

Multi-modal input

Text, image and video in one call; auto-detect input type and run appropriate classifiers.

CSAM detection

PhotoDNA + perceptual hash + classifier against NCMEC databases; 99.96% recall, <0.01% false-positive rate.

Age estimation

Performer age estimate with confidence interval; flags under-18 content for human review.

Explicit classification

Category tags (e.g. fetish, intensity, scene type), consent cues and prohibited-content flags.

2257 record helpers

Per-content record stubs with consent verification, performer IDs, content hash and timestamps.

NCMEC CyberTipline

Automated report scaffolding for legally-required CSAM reports; signed and exportable.

Action recommendations

Per-content action (allow, review, block, report) with reason codes and confidence scores.

Webhooks + audit logs

Per-decision webhooks and immutable audit logs for processor and law-enforcement evidence.

Batch + async

Submit 1,000 items in one batch; async processing with grouped webhook completion.

Process

Integration flow — step by step

Get API key

Tune thresholds

Run our sample adult dataset; tune action thresholds, false-positive tolerance and review-queue rules.

Integrate pre-publish

Wire moderation into your upload pipeline; block before content goes live.

Configure 2257 records

Map your performer IDs, consent packs and content hashes to record exports.

Set up review queue

Route borderline content to human review; integrate with your abuse-team tooling.

Go live + audit

Swap to production; export audit logs for processor conversations; quarterly threshold tuning.

GPU-optimised pipeline

Architecture

How the platform is wired

NSFW Coders engineers the full stack — from model serving and GPU orchestration to the application layer, payments, moderation and analytics. Every layer is built to be audited, scaled, and swapped without re-platforming.

Model layer

Stable Diffusion, Flux, Pony, custom LoRA & 3D — served via vLLM / ComfyUI / Triton.

API gateway

Key-auth, rate limits, credit metering, signed webhooks — multi-tenant.

Application layer

Next.js / React, realtime chat, in-app feed, creator studio.

Data & safety

Postgres + Redis + vector store, CSAM scanning, age gates, 2257 logs.

Discuss your architecture

4M+

Decisions daily

99.96%

CSAM recall

<150ms

Image p95 latency

99.9%

Uptime SLA

Tech & stack

Models & serving stack under the endpoint

Models

PhotoDNA + pHash matchersNCMEC database integrationMulti-modal classifier (NSFW-tuned)Age estimation modelConsent cue classifierCategory tagger

Serving

Triton Inference ServerRedis hash cachePostgres audit logKubernetes autoscaleA100 + L40S fleetMulti-region failover

Stack

REST + batch APIPython SDKNode.js SDKWebhook deliveryOpenAPI 3.1 specAudit log exportNCMEC report scaffolding

Use cases

What teams build on it

UGC upload moderation

Pre-publish moderation for user uploads; block prohibited content before it goes live.

Example: Platform moderating 2M uploads/month at <150ms p95.

Performer age verification

Cam-site age estimation + consent documentation at performer onboarding.

Example: Cam site onboarding 1,500 performers/month with auto-age-verify.

Marketplace category enforcement

Block banned fetish categories automatically; route borderline to review.

Example: Marketplace reducing prohibited listings by 94%.

AI CSAM attempt filtering

Detect AI-generation attempts of CSAM in user prompts and outputs.

Example: Companion app blocking 0.02% of prompts as CSAM attempts.

Processor compliance evidence

Export moderation reports during processor approval conversations.

Example: Platform passing CCBill risk review with moderation audit logs.

Hosting abuse triage

Pre-publish moderation for hosted content; route to abuse team.

Example: Adult host triaging 50k daily uploads with 3-person team.

Comparison

NSFW Coders moderation API vs alternatives

Feature	NSFW Coders	Generic moderation APIs*	Open-source self-hosted
Built for adult platforms	Classifies & documents	Refuses all adult	DIY classifier
CSAM detection	PhotoDNA + NCMEC	Varies	DIY pipeline
Age estimation	Built-in	Some providers	DIY model
2257 record helpers	Per-content exports	None	DIY
NCMEC report scaffolding	Automated	None	DIY
Multi-modal in one call	Text + image + video	Separate APIs	DIY stitching
Counsel-designed output	Processor-ready	Generic	DIY
Cost at scale	$0.004 / call	Varies	GPU capex + ops
SLA	99.9% multi-region	Provider SLA	Self-managed

* Generic moderation APIs (OpenAI, Perspective, AWS Rekognition) treat all adult content as a violation — useless when adult is your product.

Monetization

Built to make money on day one

Subscription tiers, token packs, pay-per-message, pay-per-view media, creator splits, affiliate payouts and tipping — wired into adult-friendly payment processors so revenue is never blocked by a sudden account freeze.

Subscription + credits

Hybrid billing that lifts ARPU without choking free-funnel conversion.

Creator economy

Multi-creator payouts, revenue share, content locks and PPV media.

Affiliate & referrals

Tracking links, first-touch attribution, automated payouts.

High-risk payments

Segpay, CCBill, Paxum, Verotel — with fallback routing.

Per-call moderation billing and audit log dashboard

Pricing

Moderation pricing, pay-as-you-go or dedicated

Pay-as-you-go

$0.004 / image call

Shared GPU fleet. Text at $0.001/1K tokens, video at $0.012/sec. Volume tiers drop 40–60%. Batch + async included.

Multi-modal: text + image + video
CSAM + age + 2257 + categories
NCMEC report scaffolding
Webhooks + audit logs
99.9% SLA on shared fleet

Get API access

Questions, answered

Quick answers to the questions founders ask us most about this service.