Embed grade prediction anywhere.

★ The dining room and the kitchen window

What is the CardPrepAI API?

If our website at cardprepai.com is the front door — where collectors upload card photos and get a grading prediction in their browser — then the API is the back door. Same AI brain. Same 1,000+ sources. Same 14,000+ comps. Same prediction quality. But instead of a website, your business or app talks to us through code.

Think of it like a restaurant: the dining room (website) is for diners; the kitchen window (API) is for delivery drivers and food halls. Same kitchen — different way to get the food out.

★ Six steps · two to five seconds per call

Step by step: how it works

1. Subscribe to a tier. Pick Starter, Growth, or Business based on how many predictions you need each month. Subscribe through Stripe checkout.
2. Get your API key by email. The moment your subscription activates, our system generates a unique API key and emails it to you. The key looks like cpai_live_a1b2c3d4... — store it securely.
3. Send a card to the API. From your software (Whatnot integration, card-shop POS, mobile app, script), POST card photos and your API key to cardprepai.com/api/v1/predict.
4. We do the work. Our system checks your key, verifies you're under your monthly quota, and runs the photos through the same AI that powers the website. Two to five seconds.
5. Get back a prediction. Structured data — predicted PSA grade, probability of each grade, our submit/skip recommendation, raw value estimate, PSA 10 estimate, confidence score.
6. Use it. Auto-price your Whatnot listing. Show a "predicted grade" badge to your walk-in customer. Build a feature into your app. The math is yours from there.

★ Who actually uses this

Four customer types

Marketplace sellers — Whatnot, eBay, Mercari sellers with enough volume that manually predicting each card on the website would eat their day. Best fit: Starter or Growth.

Card shops (LCS) — want walk-in customers to see an AI prediction on the spot. Powers a tablet behind the counter. Best fit: Starter.

App developers — building collection trackers, deal finders, grading-prep tools. Embed our AI instead of building one. Best fit: any tier.

Marketplaces & platforms — adding "predicted grade" as a feature for every listing. Best fit: Business or Custom.

★ Quick gut-check

Is the API right for you?

Yes — if you're running a business processing 50+ cards/month, you want predictions flowing into existing software, or you're building a product on top of CardPrepAI's brain.

No — if you're an individual collector grading 1–10 cards a week. The consumer plans (Starter $19, Pro $49, Power $149) are way more cost-effective at that volume.

Same brain. Same accuracy. Different way in.

★ Step 01 · Get your API key

Sign in, pick a tier, copy your key

API keys are shown ONCE on creation. Store them in your server-side environment. Never expose in client-side code.

# Your key looks like this:
cpai_live_a8d4e7f2c9b1...

★ Step 02 · Make a prediction call

POST a card to /api/v1/predict

curl https://cardprepai.com/api/v1/predict \
  -H "Authorization: Bearer cpai_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "front_image": "data:image/jpeg;base64,/9j/4AAQ...",
    "back_image":  "data:image/jpeg;base64,/9j/4AAQ...",
    "notes": "Pulled from a hot pack — corners look clean"
  }'

# Images must be base64 data URIs. JPEG and PNG accepted.
# back_image is optional but strongly recommended for accuracy.
# notes is optional context (max 1000 chars).

★ Step 03 · Parse the response

Structured JSON with grade probabilities + EV

{
  "ok": true,
  "elapsed_ms": 3420,
  "prediction": {
    "card_id": { "set": "2017 Panini Prizm", "card_name": "Patrick Mahomes #269 Silver" },
    "grade_prediction": {
      "psa": {
        "most_likely": "PSA 9",
        "distribution": { "10": 0.41, "9": 0.38, "8": 0.18 }
      }
    },
    "submission_recommendation": {
      "verdict": "submit",
      "best_grader": "PSA",
      "best_tier": "Value Bulk",
      "net_ev": 1820
    },
    "market_context": { "raw_nm_estimate_usd": 245, "psa_10_estimate_usd": 2150 }
  },
  "api": { "plan": "starter", "used_this_period": 47, "monthly_quota": 1000 }
}

# Response headers also include:
#   X-RateLimit-Limit:     1000
#   X-RateLimit-Used:      47
#   X-RateLimit-Remaining: 953
#   X-RateLimit-Reset:     2026-06-01T00:00:00.000Z