Batch render

Render is the simplest way to produce avatar video: pass a public audio URL plus an avatar, and the platform downloads the audio, renders the whole clip in one pass, stores the MP4, and returns its URL. No model, no realtime stream — ideal for backend jobs, batch pipelines, and pre-rendered content.

Request

POST /api/v1/render

Field	Required	Description
`audioUrl`	yes	Public URL of the speech audio to render.
`avatarId`	one of	An avatar you created (`ava_…`).
`portraitUrl`	one of	A portrait image to register on the fly.
`backgroundId`	no	Background id (defaults to plain).

Provide either avatarId or portraitUrl.

Example

import { RealtimeAvatarClient } from "realtime-avatar";

const client = RealtimeAvatarClient.platform({
  apiKey: process.env.REALTIME_AVATAR_API_KEY!,
});

const { url } = await client.render({
  audioUrl: "https://example.com/speech.mp3",
  avatarId: "ava_…",
});

console.log(url); // public MP4 URL

from realtime_avatar import RealtimeAvatarClient

client = RealtimeAvatarClient(api_key="tic_live_...")

result = client.render(
    audio_url="https://example.com/speech.mp3",
    avatar_id="ava_...",
)
print(result.url, result.duration_seconds)

curl -X POST https://realtimeavatar.ai/api/v1/render \
  -H "Authorization: Bearer $REALTIME_AVATAR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"audioUrl":"https://example.com/speech.mp3","avatarId":"ava_..."}'

Response

{
  "url": "https://.../render/rnd_….mp4",
  "avatarId": "ava_…",
  "frames": 150,
  "fps": 25,
  "durationSeconds": 6.0
}

Tip

Rendering scales with clip length, so a render call can take several seconds. Keep request timeouts generous (the SDKs default to a long timeout) and run batches concurrently.

Render bills realtime seconds against your wallet, the same as streaming turns. See Credits and billing.

Request

Example

Response

On this page