MODEL ID sync:3@0

coming-soon

sync-3

by sync. April 7, 2026

sync-3 is a lip synchronization model that processes entire shots as a single generation rather than stitching independent segments. It builds a global understanding of the speaker across all frames, enabling consistent output on close-ups, extreme face angles, partially obscured faces, and obstructed mouths. The model preserves the original speaker's style, cadence, and emotional expression across 95+ languages.

API Options

Platform-level options for task execution and delivery.

taskType string required value: videoInference: Identifier for the type of task being performed

taskUUID string required UUID v4: UUID v4 identifier for tracking tasks and matching async responses. Must be unique per task.

outputType string default: URL: Video output type.

Allowed values 1 value

outputFormat string default: MP4

Specifies the file format of the generated output. The available values depend on the task type and the specific model's capabilities.

`MP4`: Widely supported video container (H.264), recommended for general use.
`WEBM`: Optimized for web delivery.
`MOV`: QuickTime format, common in professional workflows (Apple ecosystem).

Allowed values 3 values

outputQuality integer min: 20 max: 99 default: 95: Compression quality of the output. Higher values preserve quality but increase file size.

webhookURL string URI

Specifies a webhook URL where JSON responses will be sent via HTTP POST when generation tasks complete. For batch requests with multiple results, each completed item triggers a separate webhook call as it becomes available.

Learn more 1 resource

Webhooks
PLATFORM

deliveryMethod string default: async

Determines how the API delivers task results.

Allowed values 1 value

: Returns an immediate acknowledgment with the task UUID. Poll for results using getResponse. Required for long-running tasks like video generation.

Learn more 1 resource

Task Polling
PLATFORM

uploadEndpoint string URI

Specifies a URL where the generated content will be automatically uploaded using the HTTP PUT method. The raw binary data of the media file is sent directly as the request body. For secure uploads to cloud storage, use presigned URLs that include temporary authentication credentials.

Common use cases:

Cloud storage: Upload directly to S3 buckets, Google Cloud Storage, or Azure Blob Storage using presigned URLs.
CDN integration: Upload to content delivery networks for immediate distribution.

// S3 presigned URL for secure upload
https://your-bucket.s3.amazonaws.com/generated/content.mp4?X-Amz-Signature=abc123&X-Amz-Expires=3600

// Google Cloud Storage presigned URL
https://storage.googleapis.com/your-bucket/content.jpg?X-Goog-Signature=xyz789

// Custom storage endpoint
https://storage.example.com/uploads/generated-image.jpg

The content data will be sent as the request body to the specified URL when generation is complete.

safety object

Content safety checking configuration for video generation.

Properties 2 properties

safety » checkContent checkContent boolean default: false: Enable or disable content safety checking. When enabled, defaults to fast mode.

safety » mode mode string default: none

Safety checking mode for video generation.

Allowed values 3 values

: Disables checking.
: Checks key frames.
: Checks all frames.

ttl integer min: 60: Time-to-live (TTL) in seconds for generated content. Only applies when outputType is URL.

includeCost boolean default: false: Include task cost in the response.

numberResults integer min: 1 max: 20 default: 1: Number of results to generate. Each result uses a different seed, producing variations of the same parameters.

Inputs

Input resources for the task (images, audio, etc). These must be nested inside the inputs object.

inputs » video video string required: Video input (UUID or URL).

inputs » audio audio string: Audio input (UUID or URL).

Generation Parameters

Core parameters for controlling the generated content.

model string required value: sync:lipsync@3

Identifier of the model to use for generation.

Learn more 3 resources

speech object required*

Settings for speech generation.

Properties 2 properties

speech » text text string required min: 1 max: 5000: Text to convert to speech.

speech » voice voice string required default: auto: Voice identifier to use. Set to auto for automatic selection.

Settings

Technical parameters to fine-tune the inference process. These must be nested inside the settings object.

settings » activeSpeakerDetection activeSpeakerDetection object

Speaker-targeting for multi-person clips.

Properties 4 properties

settings » activeSpeakerDetection » autoDetect autoDetect boolean default: false: Automatically detect and target the active speaker.

settings » activeSpeakerDetection » boundingBoxes boundingBoxes array of arrays: Per-frame face bounding boxes across the clip. Each box is [x1, y1, x2, y2].

settings » activeSpeakerDetection » coordinates coordinates array of numbers items: 2: [x, y] point on the target speaker's face in the selected frame.

settings » activeSpeakerDetection » frameNumber frameNumber integer min: 0: Frame index corresponding to the provided face coordinates.

settings » segments segments array of objects min items: 1

Time segments with audio sources for segmented lip sync workflows.

Properties 5 properties

settings » segments » startTime startTime float required min: 0 step: 0.01: Start time in seconds for the segment.

settings » segments » endTime endTime float required step: 0.01: End time in seconds for the segment. Must be greater than startTime.

settings » segments » audio audio string required: Audio source URL for this segment.

settings » segments » audioStartTime audioStartTime float min: 0 step: 0.01 default: 0: Start time in seconds within the source audio file.

settings » segments » audioEndTime audioEndTime float min: 0 step: 0.01: End time in seconds within the source audio file.

settings » syncMode syncMode string default: cut_off

Synchronization strategy when audio and video durations don't match.

Allowed values 4 values

: Audio bounces back and forth to fill video duration.
: Audio is cut when video ends.
: Remaining video plays with silence after audio ends.
: Audio is time-stretched or compressed to match video duration.

settings » tts tts object

Configuration for the text-to-speech provider used with speech input.

Properties 3 properties

settings » tts » provider provider string default: elevenlabs: Name of the TTS provider.

settings » tts » stability stability float min: 0 max: 1: Voice stability for the TTS provider. Higher values produce more consistent output.

settings » tts » similarityBoost similarityBoost float min: 0 max: 1: Voice similarity enforcement for the TTS provider. Higher values make the voice more closely match the target.

taskType string required value: videoInference: Type of the task.

taskUUID string required UUID v4: UUID of the task.

videoUUID string required UUID v4: UUID of the output video.

videoURL string URI: URL of the output video.

videoBase64Data string: Base64-encoded video data.

videoDataURI string URI: Data URI of the output video.

seed integer: The seed used for generation. If none was provided, shows the randomly generated seed.

NSFWContent boolean: Flag indicating if NSFW content was detected.

cost float: Task cost in USD. Present when includeCost is set to true in the request.