AI Avatar & Lip Sync Models
Turn a portrait and an audio track into a talking video, or re-sync the lips in an existing video, with OmniHuman, Kling AI Avatar, HeyGen, Sync, VEED and InfiniteTalk.
All AI Avatar & Lip Sync Models
Preview the output style, compare capabilities, then open any model for full specs, pricing, and a live Playground.
ByteDance OmniHuman 1.5
Omnihuman 1.5 is Bytedance's image-to-video AI model. Bring static images to life with fluid animation, consistent character motion, and professional-grade video output.
VEED Lipsync 2.0
VEED Lipsync 2.0 makes the person in a video articulate a new audio track.
Sync 3 Avatar Image to Video
Sync 3 animates the face in a still image so it speaks the provided audio; the output matches the audio length.
Sync 3 Lipsync
Sync 3 by sync. re-animates the speaker's lips in an existing video to match new audio, with studio-grade identity preservation.
Kling AI Avatar 2.0 Standard
Kling AI Avatar 2.0 Standard by Kuaishou animates a portrait image into a talking-head video lip-synced to an input audio track, at Standard quality.
Kling AI Avatar 2.0 Pro
Kling AI Avatar 2.0 Pro by Kuaishou animates a portrait image into a talking-head video lip-synced to an input audio track, at Pro quality.
HeyGen Avatar 4 Image to Video
Heygen Avatar4 is HeyGen's image-to-video AI model. Bring static images to life with fluid animation, consistent character motion, and professional-grade video output.
LTX 2.3 Pro Audio to Video
Ltx 2.3 Pro Audio To Video by Lightricks - advanced AI model for audio-to-video. Delivers high-quality results with fast inference, suitable for both creative and production workflows.
Wan 2.2 Speech to Video
Wan 2.2 Speech to Video turns a portrait image and a speech audio into a talking video guided by a prompt.
HeyGen Avatar 5 Digital Twin
Avatar5 Digital Twin by HeyGen - generate cinematic videos from text descriptions with AI. Create high-quality video content with natural motion, camera control, and optional audio generation.
HeyGen Lipsync Precision
HeyGen v3 Lipsync (precision) replaces the speech in a video and re-syncs the speaker's lips to the new audio.
HeyGen Lipsync Speed
HeyGen v3 Lipsync (speed) replaces the speech in a video and re-syncs the speaker's lips to the new audio.
InfiniteTalk Image to Video
InfiniteTalk turns a portrait image and speech audio into a long-form talking video with natural lip sync, head and body motion.
InfiniteTalk Text to Video
InfiniteTalk speaks the given text with a chosen voice and animates the portrait into a talking video.
InfiniteTalk Video to Video
InfiniteTalk dubs an existing video: it re-animates lips, face and body to match new speech audio.
ByteDance OmniHuman 2.0
OmniHuman 2.0 by ByteDance turns a single image and an audio track into a talking or performing video that follows the audio.
Model lineup
The most used endpoints first.
| Model | Task | What it does |
|---|---|---|
| HeyGen Avatar 4 Image to Video | Image → video | Heygen Avatar4 is HeyGen's image-to-video AI model |
| Kling AI Avatar 2.0 Pro | Image → video | Kling AI Avatar 2.0 Pro by Kuaishou animates a portrait image into a talking-head video lip-synced to an input audio track, at Pro quality |
| Kling AI Avatar 2.0 Standard | Image → video | Kling AI Avatar 2.0 Standard by Kuaishou animates a portrait image into a talking-head video lip-synced to an input audio track, at Standard quality |
| ByteDance OmniHuman 1.5 | Image → video | Omnihuman 1.5 is Bytedance's image-to-video AI model |
| VEED Lipsync 2.0 | Video → video | VEED Lipsync 2.0 makes the person in a video articulate a new audio track |
| Sync 3 Avatar Image to Video | Image → video | Sync 3 animates the face in a still image so it speaks the provided audio; the output matches the audio length |
| Sync 3 Lipsync | Video → video | Sync 3 by sync |
| LTX 2.3 Pro Audio to Video | Audio → video | Ltx 2.3 Pro Audio To Video by Lightricks - advanced AI model for audio-to-video |
| Wan 2.2 Speech to Video | Audio → video | Wan 2.2 Speech to Video turns a portrait image and a speech audio into a talking video guided by a prompt |
| HeyGen Avatar 5 Digital Twin | Text → video | Avatar5 Digital Twin by HeyGen - generate cinematic videos from text descriptions with AI |
| HeyGen Lipsync Precision | Video → video | HeyGen v3 Lipsync (precision) replaces the speech in a video and re-syncs the speaker's lips to the new audio |
| HeyGen Lipsync Speed | Video → video | HeyGen v3 Lipsync (speed) replaces the speech in a video and re-syncs the speaker's lips to the new audio |
Talking video from an image or a clip
These models do one of two jobs. Avatar models animate a still portrait so it speaks an audio track. Lip-sync models take an existing video and re-animate the speaker's lips to match new audio.
Image + audio to talking video
- OmniHuman 2.0 and 1.5 (ByteDance): a talking or performing video that follows the audio.
- Kling AI Avatar 2.0: talking-head video in Standard or Pro quality.
- HeyGen Avatar 4, Sync 3 Avatar, Wan 2.2 Speech to Video and InfiniteTalk for long-form talking video.
Re-sync an existing video
- Sync 3 Lipsync, VEED Lipsync 2.0, HeyGen Lipsync (Precision or Speed) and InfiniteTalk Video to Video for dubbing.
Need the voice first? Generate it with text-to-speech models.
Start using AI Avatar & Lip Sync Models on SandBase
One API key and one request shape for every model on this page.
- 01
Create an API key
Sign in to the SandBase Console and create a key on the API Keys page. Keep it in an environment variable and never commit it to source control.
TERMINALexport SANDBASE_API_KEY="YOUR_SANDBASE_API_KEY" - 02
Submit a run
Send a request to the Unified Run API with model bytedance/omnihuman/2.0. Only the model field changes between models on this page.
TERMINALcurl https://api.sandbase.ai/v1/run \ -H "Authorization: Bearer $SANDBASE_API_KEY" \ -H "Content-Type: application/json" \ -d '{"model": "bytedance/omnihuman/2.0", "image": "https://static.sandbase.ai/examples/4e4bda6c-b42a-40d2-8777-ec0c9dc7ca6b/input/117a23db-2541-4338-a76a-11206be9e9e7.png", "audio": "https://static.sandbase.ai/examples/4e4bda6c-b42a-40d2-8777-ec0c9dc7ca6b/input/8f429735-ac39-48b6-aa3c-360f822d11fa.mp3"}' - 03
Fetch the result
The response includes a run id. Poll it every few seconds until the status is completed, then download the output URL.
TERMINALcurl https://api.sandbase.ai/v1/run/$RUN_ID \ -H "Authorization: Bearer $SANDBASE_API_KEY"
Questions, answered
What inputs do avatar models need?
A portrait image URL and an audio URL. Some, such as InfiniteTalk Text to Video, take text and a voice instead of audio.
Avatar or lip sync?
Use an avatar model to start from a still image. Use a lip-sync model to change the speech in a video you already have.
How long can the output be?
Most models match the audio length. InfiniteTalk is built for long-form talking video.
Can I dub a video into another language?
Yes. Generate the new speech with a TTS model, then re-sync the lips with Sync 3, VEED, HeyGen Lipsync or InfiniteTalk.
Build with AI Avatar & Lip Sync Models.
One key, consistent APIs, and usage-based pricing across every model on this page.