meituan/longcat-video/distilled/image-to-video

Name: LongCat Video Distilled
Author: meituan

Longcat Video Distilled is meituan's image-to-video AI model. Bring static images to life with fluid animation, consistent character motion, and professional-grade video output.

Input

Prompt(string)

The prompt to guide the video generation.

Image URL or upload(string)

Upload

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

The URL of the image to generate a video from.

This field is required.

Number of Inference Steps(integer)

216

The number of inference steps to use. Range: 2 to 16.

Number of Refinement Inference Steps(integer)

216

The number of inference steps to use for refinement. Range: 2 to 16.

Video Output Type(string)

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

Video Quality(string)

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

Number of Frames(integer)

17961

The number of frames to generate. Range: 17 to 961.

Video Write Mode(string)

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

FPS(integer)

160

The frame rate of the generated video. Range: 1 to 60.

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 IDmeituan/longcat-video/distilled/image-to-video

Input Schema

10 parameters · 2 required · 8 optional

Parameter	Type	Required	Description
`image`	string	Required	The URL of the image to generate a video from.
`prompt`	string	Required	The prompt to guide the video generation. · Default: "First-person view from the cockpit of a Formula 1 car. The driver's gloved hands firmly grip the intricate, carbon-fiber steering wheel adorned with numerous colorful buttons and a vibrant digital display showing race data. Beyond the windshield, a sun-drenched racetrack stretches ahead, lined with cheering spectators in the grandstands. Several rival cars are visible in the distance, creating a dynamic sense of competition. The sky above is a clear, brilliant blue, reflecting the exhilarating atmosphere of a high-speed race. high resolution 4k"
`fps`	integer	Optional	The frame rate of the generated video. · Min: 1 · Max: 60 · Default: 30
`seed`	integer	Optional	The seed for the random number generator.
`num_frames`	integer	Optional	The number of frames to generate. · Min: 17 · Max: 961 · Default: 162
`video_quality`	string	Optional	The quality of the generated video. · Options: low, medium, high, maximum · Default: "high" lowmediumhighmaximum
`video_write_mode`	string	Optional	The write mode of the generated video. · Options: fast, balanced, small · Default: "balanced" fastbalancedsmall
`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)
`num_inference_steps`	integer	Optional	The number of inference steps to use. · Min: 2 · Max: 16 · Default: 12
`num_refine_inference_steps`	integer	Optional	The number of inference steps to use for refinement. · Min: 2 · Max: 16 · Default: 12

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": "meituan/longcat-video/distilled/image-to-video",
  "fps": 30,
  "image": "https://static.sandbase.ai/examples/lightricks/ltx-2.3-pro/image-to-video/input_image_0.png",
  "prompt": "First-person view from the cockpit of a Formula 1 car. The driver's gloved hands firmly grip the intricate, carbon-fiber steering wheel adorned with numerous colorful buttons and a vibrant digital display showing race data. Beyond the windshield, a sun-drenched racetrack stretches ahead, lined with cheering spectators in the grandstands. Several rival cars are visible in the distance, creating a dynamic sense of competition. The sky above is a clear, brilliant blue, reflecting the exhilarating atmosphere of a high-speed race. high resolution 4k",
  "num_frames": 162,
  "video_quality": "high",
  "video_write_mode": "balanced",
  "video_output_type": "X264 (.mp4)",
  "num_inference_steps": 12,
  "num_refine_inference_steps": 12
}'

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

← Back to LongCat Video Distilled

API README

LongCat Video Distilled

LongCat Video Distilled Image to Video animates a supplied first frame with a compact distilled workflow priced per generated second. The image anchors character appearance, composition, and scene layout, while the prompt explains how subjects, camera, atmosphere, and background elements should evolve. This route is aimed at efficient reference-led experimentation when continuity with a chosen still matters more than inventing the entire shot from text.

Its schema exposes frame count, frame rate, aspect ratio, guidance, inference steps, encoding mode, and output container so a team can tune both generation and delivery. Use it for animatics, product-motion tests, character blocking, or quick campaign variations built from approved artwork. Begin with a clean source image, describe only plausible movement, and evaluate temporal stability before increasing the length or complexity of the action.

Highlights

LongCat Video Distilled: dedicated reference-frame animation. This exact route centers on reference-frame animation; it uses prompt, image to separate its job from neighboring endpoints in the same family.
fps control. range 1 to 60; this setting provides a concrete lever for route-specific output planning.
seed control. The seed for the random number generator; use it to build controlled comparisons without changing the rest of the request.
image control. The URL of the image to generate a video from; together with the other documented fields, it supports repeatable review of this model path.

Pricing

Configuration	Price
Billing rule	`params.duration * 0.01`
Formula-priced request	Starts from `$0.010000`; final charge follows the billing rule above

When to Use

Scenario	Why this model fits
Create the exact route output	Choose it when you need image-to-video generation and already have a source image and motion direction.
Develop controlled variations	Keep the main brief fixed while changing one documented setting at a time to compare motion, framing, quality, or asset behavior.
Build repeatable batches	Use a consistent request shape for catalog, campaign, storyboard, game-asset, or social-content production.
Preserve source intent	Prefer this route when the supplied reference material must remain the foundation of a moving sequence that develops the supplied frame.
Connect a media pipeline	Use asynchronous results in an automated review, approval, post-production, or asset-management workflow.

Prompt Guide

Start with the desired result, then describe the source relationship, subject action, composition or camera behavior, lighting, style, and timing. For image-to-video generation, state what must remain stable as clearly as what should change. Use only fields exposed by the schema; the example below is structurally valid for this route.

{
  "prompt": "First-person view from the cockpit of a Formula 1 car. The driver's gloved hands firmly grip the intricate, carbon-fiber steering wheel adorned with numerous colorful buttons and a vibrant digital display showing race data. Beyond the windshield, a sun-drenched racetrack stretches ahead, lined with cheering spectators in the grandstands. Several rival cars are visible in the distance, creating a dynamic sense of competition. The sky above is a clear, brilliant blue, reflecting the exhilarating atmosphere of a high-speed race. high resolution 4k",
  "image": "https://static.sandbase.ai/examples/lightricks/ltx-2.3-pro/image-to-video/input_image_0.png",
  "fps": 30,
  "seed": 1,
  "num_frames": 162
}

Technical Specs

Specification	Value
Model ID	`meituan/longcat-video/distilled/image-to-video`
Workflow	Image-to-video generation
Required inputs	`prompt`, `image`
`fps`	`integer`; minimum: `1`; maximum: `60`
`seed`	`integer`
`image`	`string`
`prompt`	`string`
`num_frames`	`integer`; minimum: `17`; maximum: `961`
`video_quality`	`string`; options: `low`, `medium`, `high`, `maximum`
`video_write_mode`	`string`; options: `fast`, `balanced`, `small`
`video_output_type`	`string`; options: `X264 (.mp4)`, `VP9 (.webm)`, `PRORES4444 (.mov)`, `GIF (.gif)`
`num_inference_steps`	`integer`; minimum: `2`; maximum: `16`
`num_refine_inference_steps`	`integer`; minimum: `2`; maximum: `16`