For developers

An image generation API with a real design editor.

Your team designs the template on a canvas. Your code sends the data. Zandovi returns the exact image or print-ready PDF, one at a time or four hundred in a batch.

Full API on the Free plan No headless browser Deterministic output
In one paragraph

What is Zandovi?

Zandovi is a template-based image and PDF generation API with a full visual design editor. You design a template once on a canvas, bind the parts that change — text, images, QR codes, barcodes — to {{variables}}, then render personalized copies of it: one at a time by posting JSON to the REST API, or a few hundred at once as a batch job, fed by a CSV in the app or a rows array from your own code, that hands back a ZIP.

It is built for personalized assets that have to be exactly right — certificates, gift cards, vouchers, event badges, tickets and labels, where every name, code and QR must match the data. AI assistants such as Claude and ChatGPT can drive it through its MCP connector. Output is PNG, JPEG, WebP or print-ready PDF up to 4096×4096 pixels, from 200+ starter templates across 18 categories. There is a free plan and it includes the full API.

  • Not a prompt-to-image model

    Zandovi does not invent a new picture each time. You draw the design once; every render is that design with different data in it, pixel-identical everywhere else.

  • Not a headless browser

    No Chrome to keep alive, no HTML-to-PNG screenshotting. A Skia renderer draws the template server-side and streams the file back.

  • Not developer-only

    The same template is editable on a visual canvas. Designers and ops maintain it; engineers only ever send the JSON.

One template, three ways to fill it

Your team uploads a list. Your app calls the API. Your assistant does it for you.

For event, HR and school teams

Upload a spreadsheet

Drop a CSV, match columns to variables, download a ZIP. No code and no copy-paste.

How batches work

For developers whose product issues these

Call the API

One POST per image or a batch job per list, with idempotency keys and a signed webhook when it is done.

See the API

New · For anyone who works in Claude or ChatGPT

Ask your AI assistant

Connect once. Your assistant designs, checks and renders real Zandovi templates you can edit later.

Connect your assistant
Batch generation

Rows in. ZIP out.

The afternoon you used to spend changing a name and exporting, again, becomes a few minutes. Drop a CSV in the app, or submit the same rows through the API.

Read the batch generation guide Batch from the API
  1. Design once

    Build your template in the editor. Drop in shapes, text, an image, a QR.

  2. Add variables

    Wrap any text or image binding in {{double_braces}}. Set defaults and validation.

  3. Add your data

    Upload a CSV in the app — or POST the same rows to /api/v1/batch-jobs from your backend. One row per output.

  4. Get a ZIP

    PNG, JPEG or WebP at up to 4K, scale up to 4×, parallel-rendered in seconds. Poll the job, or hand it a callbackUrl and be told when it lands.

members.csv
100 rows
Menu Card · v2
3 variables
output.zip
100 PNGs · 4 MB
Rendering 100 / 100 COMPLETED
4 parallel workers · 18s elapsed Download ZIP →
Animation cycling through six real gift-card renders from the same template — recipient name, amount, message and gift code change on every frame
Six rows, one template — every frame is a real render. Only {{recipient_name}}, {{amount}}, {{gift_message}} and {{gift_code}} change.
Now live · Claude, ChatGPT and coding agents

The renderer your AI assistant calls when every name and code has to be exact.

Connect Zandovi once. Ask for a design in plain words and your assistant builds a real template, checks it against the same rules the editor uses, looks at a preview and fixes what is off. You get a link that opens it in the editor, with every changing field already a variable.

  • Checking a design is free. Only finished renders use your quota, the same pool as the API.
  • Every template keeps its last 10 versions, so a bad AI edit is one click to undo.
  • You choose what it may do, and revoke access any time in Settings → Connected apps.
claude mcp add --transport http zandovi https://mcp.zandovi.com/mcp

Works with Claude.ai, Claude Desktop, ChatGPT, Claude Code, Cursor and VS Code. Coding agents setup →

Design an Instagram coupon for my bakery's autumn sale, 25% off, code CRISP25.
  1. search_stock_images autumn pastries
  2. validate_design valid, free
  3. preview_design checked the layout
  4. create_template saved, 3 variables
Instagram coupon for the Hearth and Crumb bakery: autumn pastries photo, 25% off, promo code CRISP25

Ready for a batch or the API:

discountpromo_codeexpiry

Designed by Claude through the Zandovi connector. Not retouched.

More designs Claude made through Zandovi

See every example with its prompt
n8n community node

Add image generation to any n8n workflow.

The Zandovi node renders a template for every item that reaches it. Each template variable becomes a field you map from the step before, and the image comes out as a file the next node can email, upload or post.

  1. Install the node In n8n, open Settings → Community nodes and install the package.
    @zandovi/n8n-nodes-zandovi
  2. Connect your account Create an API key in Zandovi under Settings → API Keys and paste it into the credential.
  3. Pick a template and map its fields Choose the project and template from a list. The node shows one field per variable.
  • One field per variableLabels and default values come from the template.
  • File or linkA binary file for the next node, or a share link.
  • Batch jobsSend many items as one job and get a ZIP.
  • Charged onceA retried render is not billed twice.

Start from a ready workflow

All example workflows
An Open Graph image for a blog post with a title and an author

A share image for every new blog post

RSS feed Zandovi Save the link

Title, author and cover photo go in. A link to the finished image comes back.

Get the workflow file
A coupon with a discount, a promo code and a QR code

A coupon for every new customer

Stripe Zandovi Gmail

Each customer gets their own code and QR code, emailed as an image.

Get the workflow file
A course certificate with the recipient name and the course

A PDF certificate for every new row

Google Sheets Zandovi Gmail Drive

Add a row and the certificate is emailed and filed in Drive.

Get the workflow file

In n8n, choose Workflows → Import from File. The workflows ship without credentials; connect your own after importing.

Image generation API

One POST. One image — or four hundred.

A template-based image generation API. No headless browser, no SDK to install — a Skia-based, server-side renderer returns PNG, JPEG, WebP, or print-ready PDF up to 4096 × 4096 px at 300 DPI.

  • Per-organization API keys with usage tracking
  • Batch jobs — POST one template plus up to 400 rows, then poll or be called back, and download a ZIP
  • Signed completion callbacks — give a job a callbackUrl and it posts you a summary when it ends, HMAC-signed and retried, with the delivery status readable on the job
  • Idempotency keys on batch submits — a retried request is never billed or rendered twice, and cancelling refunds every unrendered row
  • Schema introspection — describe variables before sending
  • Sub-second renders at 1080p with Redis caching
  • Bytes by default, or a hosted URL on request — the link comes back with its exact expiry, and a revoke endpoint on every plan
Read the API docs
POST /api/v1/templates/:id/generate
curl -X POST '\
  /api/v1/templates/019e4953-7ffb/generate' \
  -H 'Content-Type: application/json' \
  -H 'X-Api-Key: YOUR_API_KEY' \
  -d '{
    "variables": {
      "price_var": "$18"
    },
    "format": "png",
    "options": {
      "outputDpi": 96,
      "quality": 90
    }
  }' \
  --output generated-image.png

# → 200 OK · image/png · streamed to disk

Real output, not mockups. Four starter templates rendered from JSON with {{variables}} — exactly what the generate endpoint returns.

API-rendered gift card with amount, recipient name and gift code filled from variables
Gift card — {{amount}}, {{gift_code}}
API-rendered Open Graph blog image with title, author and read time filled from variables
OG image — {{og_title}}, {{author_name}}
API-rendered certificate of achievement with recipient name and course filled from variables
Certificate — {{recipient_name}}, {{course_name}}
API-rendered concert ticket with attendee name, seat and scannable QR code filled from variables
Ticket — {{attendee_name}}, {{ticket_code}}
Compare

What an entry plan buys.

The cheapest paid plan of each, billed monthly, from each vendor's own pricing page.

Feature Zandovi Personal Placid APITemplate.io Templated Bannerbear Canva Bulk Create Pro
Price, billed monthly $29$19$35$39$49Part of Canva Pro
Automated renders per month 5,0005001,5001,0001,000No API
Spreadsheet batch in the app YesVia Airtable or SheetsNoYesVia integrationsUp to 300 rows
Batch API with a ZIP result YesYesYesYesYesNo
Barcode variables YesYesNoYesYesNo
Unlimited editor exports YesNoNoNoNoYes
MCP server for AI assistants YesYesNoYesYesYes
Free tier 100 renders / monthTrial credits50 / month50-credit trial30-credit trialNo bulk on Free

Checked 2026-10-17. Plans change, so check the vendor's own page for current numbers. Detailed comparisons: Zandovi vs Bannerbear, vs APITemplate.io and vs Placid, each with a cost-per-1,000-renders column and a section on when they beat us.

Render your first image in five minutes.

Create a key, pick a template, send one POST.