lightricks/ltx-2.3-22b/distilled/audio-to-video/lora

Name: LTX-2.3 22B Distilled
Author: Lightricks

Ltx 2.3 22b Distilled 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.

Input

Prompt(string)

The prompt to generate the video from.

Image URL or upload(string)

Upload

PNG, JPEG, WebP, or GIF · 20 MiB maximum

Optional URL of an image to use as the first frame of the video.

This field is required.

Image URL or upload(string)

Upload

PNG, JPEG, WebP, or GIF · 20 MiB maximum

The URL of the image to use as the end of the video.

Audio URL(string)

The URL of the audio to generate the video from.

Video Output Type(string)

The output type of the generated video. Allowed values: X264 (.mp4), VP9 (.webm), PRORES4444 (.mov), GIF (.gif).

FPS(number)

160

The frames per second of the generated video. Range: 1 to 60.

Distill Lora Second Pass Scale(number)

The scale of the distill LoRA to use for the second and subsequent passes. Range: 0 to 1.

Match Audio Length(boolean)

When enabled, the number of frames will be calculated based on the audio duration and FPS. When disabled, use the specified num_frames.

Video Quality(string)

The quality of the generated video. Allowed values: low, medium, high, maximum.

Video Size(string)

The size of the generated video. Use 'auto' to match the input image dimensions if provided. Allowed values: auto, square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9.

LoRAs(array)

The LoRAs to use for the generation.

Image Strength(number)

The strength of the image to use for the video generation. Range: 0 to 1.

Number of Frames(integer)

9481

The number of frames to generate. Range: 9 to 481.

Camera LoRA Scale(number)

The scale of the camera LoRA to use. This allows you to control the camera movement of the generated video more accurately than just prompting the model to move the camera. Range: 0 to 1.

Audio Strength(number)

Audio conditioning strength. Values below 1.0 will allow the model to change the audio, while a value of exactly 1.0 will use the input audio without modification. Range: 0 to 1.

End Image Strength(number)

The strength of the end image to use for the video generation. Range: 0 to 1.

Video Write Mode(string)

The write mode of the generated video. Allowed values: fast, balanced, small.

Use Multi-Scale(boolean)

Whether to use multi-scale generation. If True, the model will generate the video at a smaller scale first, then use the smaller video to guide the generation of a video at or above your requested size. This results in better coherence and details.

Scheduler(string)

The scheduler to use. Allowed values: ltx2, linear_quadratic, beta.

Camera LoRA(string)

The camera LoRA to use. This allows you to control the camera movement of the generated video more accurately than just prompting the model to move the camera. Allowed values: dolly_in, dolly_out, dolly_left, dolly_right, jib_up, jib_down, static, none.

Preprocess Audio(boolean)

Whether to preprocess the audio before using it as conditioning.

Seed(integer)

The seed for the random number generator.

Idle

Example output — click Run to generate your own

Production endpoint

Send your first request

OpenAI-compatible endpoint with unified authentication and usage tracking.

POSThttps://api.sandbase.ai/v1/run

Model IDlightricks/ltx-2.3-22b/distilled/audio-to-video/lora

Input Schema

22 parameters · 2 required · 20 optional

Parameter	Type	Required	Description
`image`	string	Required	Optional URL of an image to use as the first frame of the video.
`prompt`	string	Required	The prompt to generate the video from.
`fps`	number	Optional	The frames per second of the generated video. · Min: 1 · Max: 60 · Default: 24
`seed`	integer	Optional	The seed for the random number generator.
`audio`	string	Optional	The URL of the audio to generate the video from.
`loras`	object[]	Optional	The LoRAs to use for the generation.
`end_image`	string	Optional	The URL of the image to use as the end of the video.
`scheduler`	string	Optional	The scheduler to use. · Options: ltx2, linear_quadratic, beta · Default: "ltx2" ltx2linear_quadraticbeta
`num_frames`	integer	Optional	The number of frames to generate. · Min: 9 · Max: 481 · Default: 121
`video_size`	string	Optional	The size of the generated video. Use 'auto' to match the input image dimensions if provided. · Options: auto, square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9 · Default: "landscape_16_9" autosquare_hdsquareportrait_4_3portrait_16_9landscape_4_3landscape_16_9
`camera_lora`	string	Optional	The camera LoRA to use. This allows you to control the camera movement of the generated video more accurately than just prompting the model to move the camera. · Options: dolly_in, dolly_out, dolly_left, dolly_right, jib_up, jib_down, static, none · Default: "none" dolly_indolly_outdolly_leftdolly_rightjib_upjib_downstaticnone
`video_quality`	string	Optional	The quality of the generated video. · Options: low, medium, high, maximum · Default: "high" lowmediumhighmaximum
`audio_strength`	number	Optional	Audio conditioning strength. Values below 1.0 will allow the model to change the audio, while a value of exactly 1.0 will use the input audio without modification. · Min: 0 · Max: 1 · Default: 1
`image_strength`	number	Optional	The strength of the image to use for the video generation. · Min: 0 · Max: 1 · Default: 1
`use_multiscale`	boolean	Optional	Whether to use multi-scale generation. If True, the model will generate the video at a smaller scale first, then use the smaller video to guide the generation of a video at or above your requested size. This results in better coherence and details. · Default: true
`preprocess_audio`	boolean	Optional	Whether to preprocess the audio before using it as conditioning. · Default: true
`video_write_mode`	string	Optional	The write mode of the generated video. · Options: fast, balanced, small · Default: "balanced" fastbalancedsmall
`camera_lora_scale`	number	Optional	The scale of the camera LoRA to use. This allows you to control the camera movement of the generated video more accurately than just prompting the model to move the camera. · Min: 0 · Max: 1 · Default: 1
`video_output_type`	string	Optional	The output type of the generated video. · Options: X264 (.mp4), VP9 (.webm), PRORES4444 (.mov), GIF (.gif) · Default: "X264 (.mp4)" X264 (.mp4)VP9 (.webm)PRORES4444 (.mov)GIF (.gif)
`end_image_strength`	number	Optional	The strength of the end image to use for the video generation. · Min: 0 · Max: 1 · Default: 1
`match_audio_length`	boolean	Optional	When enabled, the number of frames will be calculated based on the audio duration and FPS. When disabled, use the specified num_frames. · Default: true
`distill_lora_second_pass_scale`	number	Optional	The scale of the distill LoRA to use for the second and subsequent passes. · Min: 0 · Max: 1 · Default: 0.5

Output Schema

Field	Type	Description
`id`	string	Unique identifier for the generation task
`status`	string	Task status: pending, running, completed, failed, timeout
`model`	string	Model used for the generation
`outputs`	array	Array of output items
`outputs[].url`	string	URL of the generated artifact
`outputs[].content_type`	string	MIME type (e.g. image/png, video/mp4)
`error`	object \| null	Error details if failed, null on success
`error.type`	string	Machine-readable error type code
`error.message`	string	Human-readable error description

Async Workflow

This model uses asynchronous execution. Submit a request and poll for the result.

Submit — POST to /v1/run, receive an id
Poll — GET /v1/run/{id} until status is completed, failed, or timeout
Retrieve — Read outputs from the completed response

Code Examples

Ready-to-run snippets

# Step 1: Submit
curl -X POST https://api.sandbase.ai/v1/run \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
  "model": "lightricks/ltx-2.3-22b/distilled/audio-to-video/lora",
  "fps": 24,
  "audio": "https://static.sandbase.ai/examples/lightricks/ltx-2.3-22b/distilled/audio-to-video/lora/input_audio_0.mp3",
  "image": "https://static.sandbase.ai/examples/lightricks/ltx-2.3-22b/distilled/audio-to-video/lora/input_image_1.png",
  "prompt": "A woman speaks to the camera",
  "scheduler": "ltx2",
  "num_frames": 121,
  "video_size": "landscape_16_9",
  "camera_lora": "none",
  "video_quality": "high",
  "audio_strength": 1,
  "image_strength": 1,
  "use_multiscale": true,
  "preprocess_audio": true,
  "video_write_mode": "balanced",
  "camera_lora_scale": 1,
  "video_output_type": "X264 (.mp4)",
  "end_image_strength": 1,
  "match_audio_length": true,
  "distill_lora_second_pass_scale": 0.5
}'

# Step 2: Poll result (replace <id>)
curl https://api.sandbase.ai/v1/run/<id> \
  -H "Authorization: Bearer YOUR_API_KEY"

← Back to LTX-2.3 22B Distilled

API README

LTX-2.3 22B Distilled

LTX-2.3 22B Distilled is a audio-guided video creation with custom LoRA adapters endpoint in the LTX 2.3 family. It is built for creators who need to turn a concrete creative brief into a controlled visual sequence: the request establishes the source material, the intended subject behavior, the camera language, and the atmosphere of the finished shot. The distilled, LoRA-enabled route keeps that workflow explicit instead of hiding its input assumptions behind a generic video-generation label.

In practice, this route accepts image, audio, end image, loras as creative context and exposes num frames, video size, image strength, camera lora for delivery planning. That makes it suitable for shot-based pipelines where teams must preserve a source, direct a transformation, or control the final format without losing sight of the model's central task. Write prompts as a compact shot plan—subject, action, setting, camera, light, and timing—then use the structured fields for constraints that should remain deterministic across iterations.

Highlights

Soundtrack-led performance. The audio input supplies timing and performance cues, while the prompt explains what should happen visually.
Anchored opening composition. A required opening image establishes the subject and first-frame composition before motion follows the soundtrack.
Audio influence control. audio_strength is available on this route where documented, letting a request tune adherence to the supplied sound rather than changing the prompt.
Route-specific LoRA conditioning. The loras array adds learned styling or concepts to this audio to video workflow without replacing its required source inputs. For lightricks/ltx-2.3-22b/distilled/audio-to-video/lora, this task-specific control is documented alongside the route’s own required inputs.

Pricing

Billing unit	Price
Base request	`$0.001405`

When to Use

Scenario	Recommendation
Choose this route	Use it when the deliverable specifically calls for audio-guided video creation with custom LoRA adapters, rather than a neighboring generation mode.
Prepare the source	Provide image, audio in the format described by the request schema.
Direct the shot	Describe the subject, action, environment, camera movement, lighting, and temporal progression in that order.
Control continuity	Use endpoint frames, reference media, strength, or audio controls when those fields are available instead of burying hard constraints in prose.
Plan delivery	Set duration, frame count, resolution, and aspect ratio explicitly when the schema exposes them, then compare iterations with a stable seed where supported.

Prompt Guide

For audio-guided video creation with custom LoRA adapters, describe one coherent shot rather than a list of visual keywords. Put the main subject and action first, follow with location and staging, then add camera movement, lens or framing, lighting, mood, and any timed change. Keep URLs and hard delivery choices in their dedicated fields.

{
  "prompt": "A woman speaks to the camera",
  "image": "https://static.sandbase.ai/examples/lightricks/ltx-2.3-22b/distilled/audio-to-video/lora/input_image_1.png",
  "audio": "https://static.sandbase.ai/examples/lightricks/ltx-2.3-22b/distilled/audio-to-video/lora/input_audio_0.mp3",
  "end_image": "https://example.com/reference.jpg"
}

Technical Specs

Specification	Value
Model ID	`lightricks/ltx-2.3-22b/distilled/audio-to-video/lora`
Required inputs	`prompt`, `image`
Execution	Asynchronous generation job
Request controls	22 documented fields
Output	`url`, `content_type`

Request fields

Field	Type and constraints
`fps`	number; Optional; Range: 1–60; Default: 24
`seed`	integer; Optional
`audio`	string; Optional
`image`	string; Required
`loras`	array; Optional
`prompt`	string; Required
`end_image`	string; Optional
`scheduler`	string; Optional; Options: ltx2, linear_quadratic, beta; Default: "ltx2"
`num_frames`	integer; Optional; Range: 9–481; Default: 121
`video_size`	string; Optional; Options: auto, square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9; Default: "landscape_16_9"
`camera_lora`	string; Optional; Options: dolly_in, dolly_out, dolly_left, dolly_right, jib_up, jib_down, static, none; Default: "none"
`video_quality`	string; Optional; Options: low, medium, high, maximum; Default: "high"
`audio_strength`	number; Optional; Range: 0–1; Default: 1
`image_strength`	number; Optional; Range: 0–1; Default: 1
`use_multiscale`	boolean; Optional; Default: true
`preprocess_audio`	boolean; Optional; Default: true
`video_write_mode`	string; Optional; Options: fast, balanced, small; Default: "balanced"
`camera_lora_scale`	number; Optional; Range: 0–1; Default: 1
`video_output_type`	string; Optional; Options: X264 (.mp4), VP9 (.webm), PRORES4444 (.mov), GIF (.gif); Default: "X264 (.mp4)"
`end_image_strength`	number; Optional; Range: 0–1; Default: 1
`match_audio_length`	boolean; Optional; Default: true
`distill_lora_second_pass_scale`	number; Optional; Range: 0–1; Default: 0.5