meituan/longcat-video/text-to-video/720p

Name: LongCat Video
Author: meituan

Longcat Video 720p by meituan - generate cinematic videos from text descriptions with AI. Create high-quality video content with natural motion, camera control, and optional audio generation.

Input

Prompt(string)

The prompt to guide the video generation.

Aspect Ratio(string)

The aspect ratio of the generated image. Allowed values: 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16.

Guidance Scale(number)

110

The guidance scale to use for the video generation. Range: 1 to 10.

Number of Inference Steps(integer)

850

The number of inference steps to use for the video generation. Range: 8 to 50.

Seed(integer)

The seed for the random number generator.

Video Quality(string)

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

Number of Refinement Inference Steps(integer)

850

The number of inference steps to use for refinement. Range: 8 to 50.

FPS(integer)

160

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

Video Write Mode(string)

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

Video Output Type(string)

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

Number of Frames(integer)

17961

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

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/text-to-video/720p

Input Schema

11 parameters · 1 required · 10 optional

Parameter	Type	Required	Description
`prompt`	string	Required	The prompt to guide the video generation.
`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
`aspect_ratio`	string	Optional	The aspect ratio of the generated image. · Options: 21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16 21:916:93:24:35:41:14:53:42:39:16
`video_quality`	string	Optional	The quality of the generated video. · Options: low, medium, high, maximum · Default: "high" lowmediumhighmaximum
`guidance_scale`	number	Optional	The guidance scale to use for the video generation. · Min: 1 · Max: 10 · Default: 4
`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 for the video generation. · Min: 8 · Max: 50 · Default: 40
`num_refine_inference_steps`	integer	Optional	The number of inference steps to use for refinement. · Min: 8 · Max: 50 · Default: 40

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/text-to-video/720p",
  "fps": 30,
  "prompt": "realistic filming style, a person wearing a dark helmet, a deep-colored jacket, blue jeans, and bright yellow shoes rides a skateboard along a winding mountain road. The skateboarder starts in a standing position, then gradually lowers into a crouch, extending one hand to touch the road surface while maintaining a low center of gravity to navigate a sharp curve. After completing the turn, the skateboarder rises back to a standing position and continues gliding forward. The background features lush green hills flanking both sides of the road, with distant snow-capped mountain peaks rising against a clear, bright blue sky. The camera follows closely from behind, smoothly tracking the skateboarder’s movements and capturing the dynamic scenery along the route. The scene is shot in natural daylight, highlighting the vivid outdoor environment and the skateboarder’s fluid actions.",
  "num_frames": 162,
  "video_quality": "high",
  "guidance_scale": 4,
  "video_write_mode": "balanced",
  "video_output_type": "X264 (.mp4)",
  "num_inference_steps": 40,
  "num_refine_inference_steps": 40
}'

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

← Back to LongCat Video

API README

LongCat Video

LongCat Video Text to Video 720p is the higher-resolution full-model route for generating video directly from a scene description. It offers a clearer HD canvas for evaluating faces, objects, layered movement, and environmental detail after a concept has moved beyond rough proxy testing. The endpoint is a strong fit for presentation-ready storyboards, editorial candidates, and social video where 720p is itself an acceptable delivery target.

In addition to fps, frame count, aspect ratio, guidance, inference behavior, quality, write mode, and container selection, this route provides dedicated refinement inference steps. That refinement control and the 720p raster distinguish it from the 480p path. Structure prompts around ordered visual beats, use refinement only after composition is stable, and assess both temporal continuity and fine-detail behavior before approving the final encode.

Highlights

Cinematic scene synthesis from language. LongCat Video interprets a written description as a complete moving scene, establishing subjects, environment, atmosphere, and visual progression without requiring a reference frame.
Natural motion across the generated shot. The model is designed to create high-quality video with believable movement, helping actions and environmental changes read as a connected temporal sequence rather than unrelated frames.
Camera-direction interpretation. Prompts can specify camera behavior alongside the subject action, allowing the generated scene to express deliberate viewpoint changes and cinematic movement instead of a purely static composition.
Refined 720p visual detail. The 720p route includes a dedicated refinement stage, enabling the model to revisit a generated candidate for clearer subject and environmental detail than the proxy-focused 480p workflow.

Pricing

Configuration	Price
Billing rule	`params.duration * 0.04`
Formula-priced request	Starts from `$0.040000`; 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 text-to-video generation and already have a shot description with subject, action, camera, and atmosphere.
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 coherent video interpretation of the written direction.
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 text-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": "realistic filming style, a person wearing a dark helmet, a deep-colored jacket, blue jeans, and bright yellow shoes rides a skateboard along a winding mountain road. The skateboarder starts in a standing position, then gradually lowers into a crouch, extending one hand to touch the road surface while maintaining a low center of gravity to navigate a sharp curve. After completing the turn, the skateboarder rises back to a standing position and continues gliding forward. The background features lush green hills flanking both sides of the road, with distant snow-capped mountain peaks rising against a clear, bright blue sky. The camera follows closely from behind, smoothly tracking the skateboarder’s movements and capturing the dynamic scenery along the route. The scene is shot in natural daylight, highlighting the vivid outdoor environment and the skateboarder’s fluid actions.",
  "fps": 30,
  "seed": 1,
  "num_frames": 162,
  "aspect_ratio": "21:9"
}

Technical Specs

Specification	Value
Model ID	`meituan/longcat-video/text-to-video/720p`
Workflow	Text-to-video generation
Required inputs	`prompt`
`fps`	`integer`; minimum: `1`; maximum: `60`
`seed`	`integer`
`prompt`	`string`
`num_frames`	`integer`; minimum: `17`; maximum: `961`
`aspect_ratio`	`string`; options: `21:9`, `16:9`, `3:2`, `4:3`, `5:4`, `1:1`, `4:5`, `3:4`, `2:3`, `9:16`
`video_quality`	`string`; options: `low`, `medium`, `high`, `maximum`
`guidance_scale`	`number`; minimum: `1`; maximum: `10`
`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: `8`; maximum: `50`