AI Image Rating API — Aesthetic Analysis for Model Agencies

Capabilities

Everything you get per image

Aesthetic Score

0-100 score with Low / Medium / High classification. CLIP ViT-B/16 fine-tuned on 148K rated images.

18-Dimension Aesthetic Scoring

Granular 0-10 scores across Face, Appearance, and Photography dimensions. Each dimension calibrated on real data.

Smart Tagging

7 AI-detected tags per image via Google SigLIP zero-shot classification. Portrait, fashion, editorial and more.

Content Safety

Safe / Revealing / Suggestive / NSFW classification with dedicated NSFW safety classifier cross-check and portfolio-safe flag.

Portfolio Gate

Minimal-rejection mode with learned MLP. Borderline images get a needs_review flag instead of hard rejection.

Quality Analysis

Resolution check, P99 sharpness detection, and MediaPipe face detection. Flags low-quality or faceless images for review.

Format Support

JPEG, PNG, WebP, GIF, HEIC/HEIF. Up to 25 MB. Server-side HEIC conversion included.

Face Detection

Google MediaPipe multi-scale detection returns face count, confidence, and area ratio for robust person verification.

Adaptive Learning

Submit feedback to improve AI accuracy. Hardened retraining with golden-set validation and automatic rollback.

API Reference

Endpoints

POST /api/v1/token Authenticate and get a bearer token

curl -X POST https://ai-image-rating.digitaltwin.technology/api/v1/token \
  -H "Content-Type: application/json" \
  -d '{"client_id": "agency_...", "client_secret": "sk_..."}'

POST /api/v1/rate Rate an uploaded image

Multipart file upload. Returns score, tags, content rating, quality, and portfolio gate decision.

curl -X POST https://ai-image-rating.digitaltwin.technology/api/v1/rate \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -F "file=@photo.jpg"

POST /api/v1/rate-url Rate an image from URL

curl -X POST https://ai-image-rating.digitaltwin.technology/api/v1/rate-url \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/photo.jpg"}'

GET /api/v1/usage Check your remaining quota

curl https://ai-image-rating.digitaltwin.technology/api/v1/usage \
  -H "Authorization: Bearer YOUR_TOKEN"

# {"used": 42, "limit": 1000, "remaining": 958}

POST /api/v1/feedback Submit your rating to improve the AI

curl -X POST https://ai-image-rating.digitaltwin.technology/api/v1/feedback \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"image_hash": "a1b2c3...", "user_score": 3, "dimension_feedback": {"lighting": 9.0}}'

Interactive docs available at /api/v1/docs (Swagger) and /api/v1/redoc on the API server.

Response Format

What you get back

{
  "rating": "High",
  "score": 78.5,
  "tags": [
    {"tag": "portrait", "confidence": 0.92},
    {"tag": "fashion", "confidence": 0.87},
    {"tag": "studio", "confidence": 0.65}
  ],
  "content_rating": {
    "level": "safe",
    "detail": "Professional fashion photography",
    "safe_for_portfolio": true,
    "content_rating_conflict": false
  },
  "quality": {
    "width": 3000,
    "height": 4000,
    "sharpness": 850.2,
    "issues": [],
    "is_acceptable": true
  },
  "rejected": false,
  "is_original": true,
  "portfolio_prob": 0.94,
  "image_hash": "a1b2c3d4e5f6...",
  "needs_review": false,
  "review_reason": null,
  "face_count": 1,
  "face_confidence": 0.95,
  "dimension_scores": {
    "facial_brightness": 7.4,
    "facial_clarity": 8.1,
    "facial_skin_tone": 7.9,
    "facial_structure": 7.6,
    "facial_contour": 7.2,
    "facial_aesthetic": 8.0,
    "skin_quality": 8.2,
    "hair_styling": 6.9,
    "body_proportion": 7.1,
    "clothing_fit": 8.5,
    "overall_look": 7.7,
    "color_harmony": 7.6,
    "composition": 8.0,
    "lighting": 8.4,
    "background": 7.2,
    "expression_mood": 6.5,
    "overall_cohesion": 7.8,
    "professional_quality": 8.1
  }
}

Quick Start

Up and running in 3 lines

import requests

BASE = "https://ai-image-rating.digitaltwin.technology/api/v1"

# 1. Get token
token_resp = requests.post(f"{BASE}/token", json={
    "client_id": "agency_abc123def456",
    "client_secret": "sk_your_secret_here",
})
token = token_resp.json()["access_token"]
headers = {"Authorization": f"Bearer {token}"}

# 2. Rate an image
with open("photo.jpg", "rb") as f:
    resp = requests.post(f"{BASE}/rate", headers=headers, files={"file": f})

result = resp.json()
print(f"Score: {result['score']}, Rating: {result['rating']}")
print(f"Tags: {[t['tag'] for t in result['tags']]}")
print(f"Portfolio suitable: {not result['rejected']}")

const BASE = "https://ai-image-rating.digitaltwin.technology/api/v1";

// 1. Get token
const tokenResp = await fetch(`${BASE}/token`, {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({
    client_id: "agency_abc123def456",
    client_secret: "sk_your_secret_here",
  }),
});
const { access_token } = await tokenResp.json();

// 2. Rate an image from URL
const resp = await fetch(`${BASE}/rate-url`, {
  method: "POST",
  headers: {
    Authorization: `Bearer ${access_token}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ url: "https://example.com/photo.jpg" }),
});

const result = await resp.json();
console.log(`Score: ${result.score}, Rating: ${result.rating}`);

# 1. Get token
TOKEN=$(curl -s -X POST https://ai-image-rating.digitaltwin.technology/api/v1/token \
  -H "Content-Type: application/json" \
  -d '{"client_id": "agency_...", "client_secret": "sk_..."}' \
  | python3 -c "import sys,json; print(json.load(sys.stdin)['access_token'])")

# 2. Rate an image file
curl -X POST https://ai-image-rating.digitaltwin.technology/api/v1/rate \
  -H "Authorization: Bearer $TOKEN" \
  -F "file=@photo.jpg"

# 3. Rate from URL
curl -X POST https://ai-image-rating.digitaltwin.technology/api/v1/rate-url \
  -H "Authorization: Bearer $TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/photo.jpg"}'

# 4. Check usage
curl https://ai-image-rating.digitaltwin.technology/api/v1/usage \
  -H "Authorization: Bearer $TOKEN"

Pricing

Start free. Scale when ready.

Starter

Free

1,000 requests / day
All endpoints included
OAuth2 + JWT authentication
Interactive API docs

Request Access

Popular

Enterprise

Custom

Unlimited requests
Priority GPU workers
Per-agency AI training
Dedicated support + SLA

Contact Sales

Get Started

Request API access

We review every application. You'll receive credentials once approved.

Already applied?

Aesthetic image analysis, built for model agencies