easyrender

Turn any Product Photo into a 3D model

Package Information

Released: this month
Downloads: 0 weekly / 0 monthly
Latest Version: 0.2.0
Author: easyrender

Documentation

n8n-nodes-easyrender

Connect easyrender product photos and 3D model workflows to n8n. This is the
source package for a community node submission; it is not yet published or verified.

Requirements and credentials

Use Node.js 22 or later for development. In easyrender Connectors,
create an er_ API key with read and write scopes. In n8n, create an
Easyrender API credential and paste only the key. The credential test reads
GET /api/credits; it does not submit a render. Read-only workflows need only read.
The key is sent as Authorization: Bearer … to https://easyrender.ai/api.

Deployment prerequisite: deploy this PR's REST image_url support before using
Create Render. See the API contract. No base URL override is exposed.

Operations

Node / operation Behavior
easyrender / Create Render Creates a private preview from a public HTTPS image URL. Requires a stable idempotency key.
easyrender / Get Render Reads status, downloads available images into Binary, and adds viewer/model links.
easyrender / List Renders Returns one envelope with renders, balance, nextCursor; retain filters and pass that cursor to fetch another page.
easyrender / Get Credits Returns current account usage information.
easyrender Trigger / Model Ready Polls one configured render ID. Emits only status=ready and hasModel=true.

Creation starts a preview. Review the preview and angle sheet, then start the 3D build
separately in easyrender. No spend or share scope is needed by these nodes.
Images must be public HTTPS JPEG, PNG or WebP, at most 4 MiB. The easyrender server
fetches the photo through its protected image loader; n8n never sends the API key to the image host.

Map a stable source event ID into Idempotency Key (1–128 printable ASCII characters,
no spaces). Reuse the same key and exact image/input after a lost response. Do not
retry with a fresh key or silently change the image at the same URL.

See the result in n8n

  1. Create Render starts a preview and returns an id. It does not start the 3D build.
  2. After a Wait step, use Get Render with that ID and Download Images enabled
    (the default). Open the node's Output → Binary tab, then View on preview
    or sheet to inspect the image. PNG, JPEG and WebP keep their response MIME type
    and matching filename extension.
  3. Open viewerUrl, sign in as the owner for a private render, and create/review the
    angle sheet before starting Build 3D in easyrender.
  4. Run Get Render again, or use Model Ready in a separate workflow with the fixed
    render ID. When ready, inspect the images in Binary and use the model links in JSON.

Both Get Render and Model Ready preserve all existing JSON fields and add:

Field Meaning
viewerUrl https://easyrender.ai/r/<id>; public renders are viewable publicly, private renders require owner access. This does not publish a render.
glbUrl Model download endpoint when status=ready and hasModel=true; otherwise null. Private downloads require the same bearer API key (or an owner browser session).
usdzUrl null: USDZ is converted in the browser, with no server download URL. Open the viewer and choose USDZ to export it.
summary A readable status and next step, such as Ready - 3D model available.

The API's /preview endpoint returns the photo before a sheet exists, and the angle
sheet once available. binary.preview contains that current preview; when hasSheet
is true, binary.sheet also exposes the sheet with its own filename. Both refer to
the same downloaded asset, fetched once. Private image requests use your Easyrender API
credential; credentials are never put in output links or forwarded through redirects.
Missing images (absent flags, HTTP 404 or 410) leave JSON intact without binary data.
Other download errors are surfaced so authentication and service failures remain visible.
Disable Download Images to return only JSON and links.

Import the example workflow JSON, select your Easyrender
API credential in both Easy Render nodes, and replace the example image URL and
idempotency key. It creates a preview, waits, and reads the result with images enabled.
The wait is not a completion guarantee: if still processing, run Get Render again with
that ID. Review and start the 3D build in the viewer, then run Get Render for model links.
For production automation, use a bounded wait/poll loop with ready/failed checks.

To download a private GLB within n8n, add an HTTP Request node: GET {{$json.glbUrl}},
Authentication → Predefined Credential Type → Easyrender API, Response Format → File.
Images are already present in the Easy Render output; a GLB remains a file, not an
interactive 3D viewport inside n8n.

Example workflows

  1. A product webhook provides image_url, event_id and name.
  2. Add easyrender → Create Render; map those fields to Image URL, Idempotency Key and Name.
  3. Save the returned id with the source product.
  4. Use Get Render after a Wait node to check the preview. Review and start the build in easyrender.
  5. In a separate active workflow, configure easyrender Trigger with that render ID,
    a polling interval of a minute or longer, and a downstream delivery step.

The trigger watches a single configured ID, not all account renders. n8n workflow
static data suppresses repeats after a successful active execution. Manual tests return
an already-ready model without consuming the event. For a different render, use a new
trigger instance. Failed jobs raise an error. Workflows may be retried by the host,
so downstream delivery should also use the render ID for deduplication.

For dynamic pipelines, use scheduled Get Render steps for IDs saved by Create Render,
with a ready/failed condition and a bounded wait. The trigger cannot take an upstream
item because it is the workflow's starting node. An account-wide completion feed is
not promised: the API lists by creation time, not completion time.

Private model downloads require owner authentication at /api/renders/{id}/model;
passing that URL to a third party does not grant access. These nodes do not publish models.

Errors

401: reconnect; 403: check scopes; 404: check ID and ownership; 409: reuse the original
idempotency key with the original input; 429: wait at least Retry-After before a retry.
The HTTP helper surfaces API failures. No automatic retries are implemented in this package.
Inspect an uncertain job by ID before retrying a mutation; preserve request_id for support.

Local development and validation

cd integrations/n8n
npm ci
npm run lint
npm run build
npm test
npm pack --dry-run
npm run dev

The official @n8n/node-cli handles build, icon copying, local development and linting.
npm test builds and exercises the actual node methods with a fake n8n transport; it
never contacts production. For a hosted/editor smoke test, use the example above with
an authorized test key. Source follows the n8n starter.
The package has no runtime dependencies, environment access or filesystem access in its nodes.

After publication, self-hosted users can install n8n-nodes-easyrender via Settings →
Community Nodes. Cloud discovery requires n8n verification; do not claim that status
until approved. See listing and release instructions.

Compatibility and support

Built against n8n-workflow 2.16.0 with node API version 1. The wildcard peer follows
n8n's starter convention; it is not evidence of testing every older n8n release.
Support: GitHub issues.
Licensed under MIT.

Discussion