IMAGE ANALYSIS API

Turn images into structured JSON

The image analysis API for developers. POST a photo or share a capture link. Define the fields you need, get objects, text, and conditions back as JSON in seconds.

No credit card required. 50 free credits to start.
POST /v1/images
{ "object": "sneaker", "brand": "nike", "condition": "used", "confidence": 0.95 }
200 OK · 1.4s

HOW IT WORKS

Image in, JSON out. Three steps.

01

Define your schema

List the fields you want back, like object, brand, damage type, or text on the image. You choose them, the API fills them in.

02

Send the photo

POST an image directly, or generate a capture link. Users take a photo or upload from any browser, no app needed.

03

Get your JSON

A webhook fires the moment analysis finishes, with every field you defined already filled in.

ONE SCHEMA ENGINE

What you can pull out of an image

Define the fields you need. The API returns exactly those, structured as JSON, for every photo.

UNIFIED SCHEMA

Define your fields once. Extract them from every photo.

{ object: string, condition: enum, defect_type: string }
OBJ
Objects

Products, vehicles, equipment, documents, people.

TXT
Text

Visible text, on-image labels, error codes, signs.

CND
Conditions

Damage type, severity, quality, wear, defects.

#
Counts

People, items, occurrences, missing pieces.

HEX
Colors

Specific colors, dominant palette, color matching.

BRD
Brands and logos

Logo detection, brand identification.

LYT
Layout

Document structure, form fields, regions of interest.

+
Custom fields

Define a field, get it back.

USE CASES

What teams build with the image analysis API

E-commerce returns

Customer snaps the damaged item. Extract defect type, condition, packaging state.

{ defect_type: "cracked_case", packaging_state: "damaged" }
Insurance claims

Claimant photographs the damage. Extract damage type, severity, location on vehicle.

{ damage_type: "dent", severity: "minor" }
Identity verification

ID cards and passports captured via link. Extract name, document number, expiry date.

{ document_type: "passport", expiry_date: "2029-04-01" }
Receipts and invoices

Snap a receipt, get back vendor, line items, total, tax.

{ vendor: "Acme Supplies", total: 84.20 }
Manufacturing defects

QA photos from the line. Extract defect type, location, pass or fail.

{ defect_type: "scratch", result: "fail" }
AI agents with eyes

Drop the API into your agent toolkit. Let users show problems instead of describing them.

{ object: "leaking_pipe", confidence: 0.91 }

PRICING

Start free. Scale when you need to.

Pay as you go
50 credits free
then $1 per 10 credits
  • 1 credit per image
  • Files auto-delete after 1 hour
  • Up to 3 assistants for different extraction tasks
  • Custom fields on every request
  • Capture links, up to 10 images each
Start free
Pro
$149/mo
2,000 credits included
  • Everything in Pay as you go
  • Shared workspace for up to 3 teammates
  • Unlimited assistants
  • Fine-tune extraction with your own data
  • Custom branding on capture links
  • Handwriting OCR included
Start free
Enterprise
Custom
for teams with scale or compliance needs
  • On-premise deployment
  • Custom-trained extraction models
  • Volume discounts
  • Priority support
Contact sales

FAQ

Questions, answered

How is this different from an OCR API?

OCR gives you raw text. We give you structured data. Define a schema, and the API returns the extracted text plus the fields you actually need: object detection, condition, severity, counts.

What image formats do you support?

JPEG, PNG, WEBP, HEIC, TIFF, BMP, and PDF (per-page extraction). Up to 25 MB per file on Pay as you go, larger on Pro and Enterprise.

Can you extract text from images?

Yes. Add a text_on_image field to your schema and you get back every visible string. Handwriting recognition is included on Pro and above.

Can you detect specific objects or brands?

Yes. The model recognizes thousands of common objects and brand marks out of the box. For your own SKUs or custom defect types, fine-tune with your own data on Pro.

Can you compare two images?

Yes. Pass two image URLs and a field like match_score in your schema, and the API returns a similarity score with reasoning.

Can users take photos in the browser?

Yes. The capture link opens any device camera through the browser, no app needed. Works on iOS Safari, Android Chrome, and desktop browsers.

Start analyzing images today

50 free credits. No credit card required.

Start free