InfiniteTalk API
Audio-driven talking-head animation — lip-sync and animate a portrait from any audio input.
curl -X POST https://runapi.ai/api/v1/infinitetalk/audio_to_video \
-H "Authorization: Bearer $RUNAPI_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "infinitetalk-from-audio",
"image_url": "https://cdn.runapi.ai/public/samples/portrait.jpg",
"audio_url": "https://cdn.runapi.ai/public/samples/voice.mp3"
}'import { InfinitetalkClient } from "@runapi.ai/infinitetalk";
const client = new InfinitetalkClient();
const result = await client.audioToVideo.run({
model: "infinitetalk-from-audio",
image_url: "https://cdn.runapi.ai/public/samples/portrait.jpg",
audio_url: "https://cdn.runapi.ai/public/samples/voice.mp3",
});<?php
require __DIR__ . "/vendor/autoload.php";
use RunApi\Infinitetalk\InfinitetalkClient;
$client = new InfinitetalkClient();
$result = $client->audioToVideo->run([
'model' => 'infinitetalk-from-audio',
'image_url' => 'https://cdn.runapi.ai/public/samples/portrait.jpg',
'audio_url' => 'https://cdn.runapi.ai/public/samples/voice.mp3',
]);require "runapi/infinitetalk"
client = RunApi::Infinitetalk::Client.new
result = client.audio_to_video.run(
model: "infinitetalk-from-audio",
image_url: "https://cdn.runapi.ai/public/samples/portrait.jpg",
audio_url: "https://cdn.runapi.ai/public/samples/voice.mp3"
)npx skills add runapi-ai/infinitetalk -g# Claude Code
claude mcp add runapi -s user -- npx -y @runapi.ai/mcp
# Codex
codex plugin install runapi-mcp@agents
# Cursor / Windsurf / VS Code
npx @runapi.ai/mcp init cursorInfiniteTalk is MeiGen-AI's audio-driven talking head animation model that generates lip-synced facial animations from an audio input and a reference image. It produces natural head movements and expressions synchronized to speech.
- 多個變體，對應不同速度／品質層級
- Model skill 包含文件、schema 與 setup 備註
- 支援 app-focused coding workflows
- 生成失敗不收費
收費
技術詳情
| 模型 ID | infinitetalk-from-audio |
| 供應商 | MeiGen-AI |
| 模式 | video |
| 任務類型 | asynchronous |
| 計費單位 | second |
| API endpoint | /api/v1/infinitetalk/audio_to_video |
| 商業授權 | 是 — 已透過 API 包含 |
| 目錄狀態 | 可正常運作 |
為 app 開發安裝 InfiniteTalk skill
將模型文件、schema、價格備註與 setup 步驟載入 coding workspace。
# Install the model skill for app development workflows
npx skills add runapi-ai/infinitetalk -gInstall the InfiniteTalk skill for this app: 1. Add runapi-ai/infinitetalk with the skills installer. 2. Load SKILL.md in this workspace. 3. Use its docs, schemas, pricing notes, and setup steps when adding model features. 4. Confirm the install path when done.
從 model skill 到第一次結果，只需四步
選擇模型
挑選符合輸出類型、品質門檻與延遲目標的模型與變體。
設定
設定 RunAPI key，並在 coding workspace 安裝 model skill。
開發
使用 skill 指引，在你的 app 內加入模型功能。
接收
透過 task ID 查詢、在支援時串流，或處理 webhook callback。
InfiniteTalk API 是甚麼？
InfiniteTalk is MeiGen-AI's audio-driven avatar model, generating lip-synced animation from a portrait and audio input. Through RunAPI, it shares unified API access and billing.
為何透過 RunAPI 使用 InfiniteTalk API
一個驗證，對應所有 provider
一組 RunAPI key 即可解鎖整個 catalog。無需分開帳戶，亦無需為每個整合輪換密鑰。
統一定價與計費
按次以 USD 收費，每月結算。失敗的生成不會收費。
內含 schema 的 skill
型別化 schema 與 setup 備註打包在 model skill 內，讓實作從正確契約開始。
常見問題
What inputs does InfiniteTalk require?
A portrait image (face clearly visible) and an audio file. The model generates a video of the portrait speaking the audio with lip sync.
How accurate is the lip sync?
InfiniteTalk produces frame-accurate lip sync matched to the input audio, with natural mouth shapes for speech sounds.
Does it animate more than just the mouth?
Yes — the model generates natural head movement, eyebrow expressions, and subtle facial micro-expressions synchronized with speech patterns.
What audio formats are supported?
MP3, WAV, AAC, MP4, and OGG are supported formats. The audio drives both lip sync timing and expression intensity.
Is there a maximum video length?
InfiniteTalk supports generating long-form talking head video — the output length matches the input audio duration.
我應該先用邊個版本？
先選最平而且符合你質素要求的版本。大多數團隊會先用快速版本，之後再升級到 pro 用於正式生產。
有免費方案嗎？
新帳戶可獲得每個模型的首次呼叫免費。之後按次收費。
你哋支援串流結果嗎？
如有提供串流，RunAPI 會端到端串流。
失敗會點樣收費？
生成失敗不會收費。
輸出有快取嗎？
已生成輸出會儲存，並可按任務 ID 取回。輸入內容不會快取。
可以商業使用嗎？
可以——除非模型授權明確限制，否則每個版本都包括商業使用。相關限制會在版本頁面列明。
限流點樣計？
每個密鑰的限流會跟使用等級而提升。最新限制請參考定價頁。
可以在邊度報告問題？
可以在公開 GitHub repo 開 issue，或者電郵支援。
如果你鍾意 InfiniteTalk API，可以試吓呢些
開始用 InfiniteTalk API 開發。
- audio_to_video