For AI agents: visit https://elai.readme.io/llms.txt for an index of all pages formatted in Markdown and endpoints in OpenAPI. Append .md to any documentation page URL to get its markdown version.
Render an existing video that was created before using another request, or with Elai Builder. It can also trigger the re-rendering of failed videos of previous batch requests.
❗️
This endpoint accepts work, it does not complete it
A 200 means the render was queued. Validation runs after the response is sent, so no render
failure can reach you as an HTTP status. Failures arrive as a video_error webhook, or as status: "error" on GET /videos/{videoId} if you poll.
Messages you may receive that way:
Message
Cause
Slide N is too long. Please limit speech to Xs.
Speech exceeds the global per-slide limit
Slide N is too long for this avatar. Please limit speech to Xs.
Speech exceeds this avatar's own limit, which is lower
Slide N uses a premium avatar that is no longer available...
The slide's premium avatar is not available to your plan
You do not have enough minutes to render this video...
No minutes left, and the workspace is not allowed to overrun
Your video has N slides but your plan allows only 3 slides to render.
Free plans render at most 3 slides
Video with music should be minimum 2 seconds long
Background music on a video shorter than 2s
If you cannot afford to miss a video_error — deliveries are single-attempt, with no retry —
reconcile with GET /videos instead of relying on the callback alone.
Recent Requests
Log in to see full request history
Time
Status
User Agent
Retrieving recent requests…
Loading…
Render existing video, that was created before using another request or with Elai Builder. Also, it can trigger the re-render of failed videos of previous batch requests.
Responses
403
The video belongs to another workspace.
404
The video does not exist, or the id is not a valid ObjectId.
Updated 18 days ago
Did this page help you?
Yes
No
Language
ShellNodeRubyPHPPython
Credentials
Header
Header
Loading…
Response
Click Try It! to start a request and see the response here! Or choose an example: