Documentation

Renderly for developers

Renderly turns HTML (or a saved template) into print-ready documents — PDF or PNG — through a single API call. This documentation covers every detail: authentication, the render endpoint, templates, variables, verifiable documents, errors and limits.

If you have 5 minutes, start here. The steps below take you from zero to your first PDF — and the rest of the documentation digs into each concept.

What Renderly does

Generating documents on the server is usually painful: PDF libraries that do not understand modern CSS, headless browsers to maintain, print layouts that break. Renderly solves this as a service. You describe the document in HTML — or build a template in the visual editor — and we return the rendered file, faithful to what you designed.

Real HTML, real rendering

Raw HTML with variable substitution, or a saved template that compiles tables, QR codes and charts. Your CSS behaves just like in a browser.

Two paths, one route

Send html (raw HTML) or template (a saved slug). The same endpoint resolves both. You never maintain a headless browser.

Verifiable documents

Every render produces a SHA-256 of the bytes and a public /verify page. Tamper-proof integrity, built in.

Synchronous and direct

The request response is the binary itself (PDF or PNG). No polling, no queue for the common case.

Your first 5 minutes

Three steps, from zero to a file on your disk.

1. Get your API key

In the Dashboard you generate a key in the format rndr_live_.... It appears only once — copy it and store it in a secrets vault.

2. Make your first call

Send a POST to /v1/render with a template and your data. The PDF comes back in the response — in ~10ms via the native engine.

3. Download the binary

The response is the file. With curl, use --output; in code, read the body as bytes and save it. Done.

In practice, your first call looks exactly like this:

POST /v1/render — your first PDF

curl -X POST https://api.renderly.dev/v1/render \
  -H "Authorization: Bearer rndr_live_SEU_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "html": "<h1>Olá, {{nome}}</h1><p>Seu primeiro PDF pela Renderly.</p>",
    "data": { "nome": "Maria" },
    "format": "pdf"
  }' --output ola.pdf

The secret of {{nome}}

The html has a {{nome}} and the data carries { "nome": "Maria" }. Renderly substitutes the variable before rendering. It is the same variable syntax used in templates — see Data & variables.

How rendering works, under the hood

Every call goes through the same pipeline, and understanding it helps you debug and optimize:

1Authentication. Your API key is validated and mapped to an organization.
2Quota. We check how many renders the organization has already made in the current month against the plan limit.
3Resolution. We decide what to render: a saved template (compiled with your data) or the raw HTML you sent.
4Render. A two-track router picks the engine: native (deterministic, no browser) for compatible documents, or Chromium for raw HTML, PNG and complex layouts.
5Issuance. We compute the SHA-256 of the bytes, log the issuance for verification, and respond with the file + diagnostic headers.

Two engines, one response

The X-Render-Engine response header tells you whether that document came out of the native engine or chromium. You do not choose — the router decides based on the document structure. Both produce the same output contract.