LongCat-Video t2v API

meituan-longcat/longcat-video-t2v-480p

Turn plain text into cinematic, on-brand video. Describe the scene and camera feel; LongCat generates smooth, consistent shots with adjustable length, FPS, and quality.

0.02 per second of output video

Input

Prompt

Realistic, cinematic filming style. An advanced skateboarder in a dark full-face helmet, deep-colored windbreaker, blue jeans, and bright yellow shoes rides along a winding alpine mountain road. Setting: lush green hills on both sides, distant snow-capped peaks, clear bright blue sky, crisp natural daylight. Shot duration ~8–10 seconds, single continuous take, no cuts.

Action beats: at start the rider is already rolling; camera tracks closely from behind at low height (~1.5 m). The rider accelerates into an S-curve, then gradually lowers into a crouch; at mid-shot they extend the right hand to lightly touch the asphalt while keeping a low center of gravity to carve a sharp left hairpin. After completing the turn, the rider rises smoothly back to a standing position and continues gliding forward with relaxed balance.

Camera direction: stabilized follow shot on a gimbal/drone/pursuit rig ~2–3 m behind, slight parallax to the road edge; gentle ease-in/ease-out speed changes; subtle lateral drift to reveal scenery; horizon level (no Dutch angles). Lens ~35 mm equivalent, medium depth of field; natural motion blur at 24–30 fps. Keep framing center-weighted with road lines leading forward; maintain ~60% ground, ~40% sky.

Lighting & mood: bright natural daylight with soft bounce; clean contrast, saturated greens, blue sky, white peak highlights; realistic color and textures. Physical realism: board trucks tilt into the turn, wheels grip the road, clothing and shoelaces react to wind, slight dust wisp where the hand grazes the surface.

Avoid: jerky camera moves, jitter, rubbery or warped limbs, melting board/trucks, exaggerated speed ramps, floating shadows, text or watermarks. Keep everything coherent, grounded, and on-brand, delivering a smooth, premium commercial feel.

The prompt to guide the video generation.

Number of Frames

Min: 17 - Max: 961

The number of frames to generate.

Video Output Type

The output type of the generated video.

Video Write Mode

The write mode of the generated video.

Aspect Ratio

The aspect ratio of the generated video.

FPS

Min: 1 - Max: 60

The frame rate of the generated video.

Video Quality

The quality of the generated video.

Enable Safety Checker

Safety checker can only be disabled on API call

Number of Inference Steps

Min: 2 - Max: 16

The number of inference steps to use.

Seed

The seed for the random number generator.

You need to be logged in to run this model and view results.

Output

{
  "error": "",
  "inferenceTime": 0,
  "output": "https://media.modelrunner.ai/zRoAzFbTjpDMmdP2pTZn5_tmpz44mxh1i.mp4",
  "input": {
    "fps": 15,
    "seed": 0,
    "prompt": "Realistic, cinematic filming style. An advanced skateboarder in a dark full-face helmet, deep-colored windbreaker, blue jeans, and bright yellow shoes rides along a winding alpine mountain road. Setting: lush green hills on both sides, distant snow-capped peaks, clear bright blue sky, crisp natural daylight. Shot duration ~8–10 seconds, single continuous take, no cuts.\n\nAction beats: at start the rider is already rolling; camera tracks closely from behind at low height (~1.5 m). The rider accelerates into an S-curve, then gradually lowers into a crouch; at mid-shot they extend the right hand to lightly touch the asphalt while keeping a low center of gravity to carve a sharp left hairpin. After completing the turn, the rider rises smoothly back to a standing position and continues gliding forward with relaxed balance.\n\nCamera direction: stabilized follow shot on a gimbal/drone/pursuit rig ~2–3 m behind, slight parallax to the road edge; gentle ease-in/ease-out speed changes; subtle lateral drift to reveal scenery; horizon level (no Dutch angles). Lens ~35 mm equivalent, medium depth of field; natural motion blur at 24–30 fps. Keep framing center-weighted with road lines leading forward; maintain ~60% ground, ~40% sky.\n\nLighting & mood: bright natural daylight with soft bounce; clean contrast, saturated greens, blue sky, white peak highlights; realistic color and textures. Physical realism: board trucks tilt into the turn, wheels grip the road, clothing and shoelaces react to wind, slight dust wisp where the hand grazes the surface.\n\nAvoid: jerky camera moves, jitter, rubbery or warped limbs, melting board/trucks, exaggerated speed ramps, floating shadows, text or watermarks. Keep everything coherent, grounded, and on-brand, delivering a smooth, premium commercial feel.",
    "num_frames": 162,
    "aspect_ratio": "16:9",
    "video_quality": "high",
    "video_write_mode": "balanced",
    "video_output_type": "X264 (.mp4)",
    "num_inference_steps": 15,
    "enable_safety_checker": false
  },
  "logs": "Generated video (undefinedxundefined, application/octet-stream)"
}

Logs (1 lines)

Examples

LongCat-Video t2v API

LongCat-Video t2v is a text-to-video AI model by meituan-longcat. On ModelRunner it runs through a REST API or via MCP from any AI assistant, at $0.02 per second of video.

POST https://queue.modelrunner.run/meituan-longcat/longcat-video-t2v-480p

cURL

# Submit a request to the queue. Input fields go at the top level of the
# body. The optional reserved "metadata" object holds your own flat string
# tags — stored on the request, never sent to the model; filter later with
# GET https://queue.modelrunner.run/requests?metadata=<url-encoded JSON>.
curl -X POST https://queue.modelrunner.run/meituan-longcat/longcat-video-t2v-480p \
  -H "Authorization: Key $MRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "fps": 15,
    "seed": 0,
    "prompt": "Realistic, cinematic filming style. An advanced skateboarder in a dark full-face helmet, deep-colored windbreaker, bl…",
    "num_frames": 162,
    "aspect_ratio": "16:9",
    "video_quality": "high",
    "video_write_mode": "balanced",
    "video_output_type": "X264 (.mp4)",
    "num_inference_steps": 15,
    "enable_safety_checker": false,
    "metadata": {
      "project": "my-project"
    }
  }'
# → { "request_id": "...", "status_url": "...", "response_url": "..." }

# Poll status_url until "COMPLETED", then fetch the result
curl "https://queue.modelrunner.run/meituan-longcat/longcat-video-t2v-480p/requests/$REQUEST_ID/status" \
  -H "Authorization: Key $MRUN_API_KEY"
curl "https://queue.modelrunner.run/meituan-longcat/longcat-video-t2v-480p/requests/$REQUEST_ID" \
  -H "Authorization: Key $MRUN_API_KEY"

JavaScript

import { modelrunner } from "@modelrunner/client";

const result = await modelrunner.subscribe("meituan-longcat/longcat-video-t2v-480p", {
  input: {
    "fps": 15,
    "seed": 0,
    "prompt": "Realistic, cinematic filming style. An advanced skateboarder in a dark full-face helmet, deep-colored windbreaker, bl…",
    "num_frames": 162,
    "aspect_ratio": "16:9",
    "video_quality": "high",
    "video_write_mode": "balanced",
    "video_output_type": "X264 (.mp4)",
    "num_inference_steps": 15,
    "enable_safety_checker": false
  },
});
console.log(result);

Python

import os
import requests

headers = {"Authorization": f"Key {os.environ['MRUN_API_KEY']}"}

submitted = requests.post(
    "https://queue.modelrunner.run/meituan-longcat/longcat-video-t2v-480p",
    headers=headers,
    json={
      "fps": 15,
      "seed": 0,
      "prompt": "Realistic, cinematic filming style. An advanced skateboarder in a dark full-face helmet, deep-colored windbreaker, bl…",
      "num_frames": 162,
      "aspect_ratio": "16:9",
      "video_quality": "high",
      "video_write_mode": "balanced",
      "video_output_type": "X264 (.mp4)",
      "num_inference_steps": 15,
      "enable_safety_checker": false
    },
).json()

# Poll submitted["status_url"] until "COMPLETED", then:
result = requests.get(submitted["response_url"], headers=headers).json()

Input parameters

Name	Type	Required	Description
prompt	string	yes	The prompt to guide the video generation.
num_frames	integer	no	The number of frames to generate. Default: 162.
video_output_type	enum	no	The output type of the generated video. Default: "X264 (.mp4)".
video_write_mode	enum	no	The write mode of the generated video. Default: "balanced".
aspect_ratio	enum	no	The aspect ratio of the generated video. Default: "16:9".
fps	integer	no	The frame rate of the generated video. Default: 15.
video_quality	enum	no	The quality of the generated video. Default: "high".
enable_safety_checker	boolean	no	Whether to enable safety checker. Can't be disabled in web UI. Default: true.
num_inference_steps	integer	no	The number of inference steps to use. Default: 12.
seed	integer	no	The seed for the random number generator. Default: 0.

Machine-readable: OpenAPI schema · llms.txt

Use LongCat-Video t2v from Claude & Cursor (MCP)

Point Claude Code, Claude Desktop, Cursor, or any MCP client at the ModelRunner MCP server and LongCat-Video t2v becomes a tool your assistant can call directly — it authorizes via OAuth (no API key in config) and runs this model with the run_model tool using the endpoint meituan-longcat/longcat-video-t2v-480p.

MCP client config (Claude Desktop, Cursor)

{
  "mcpServers": {
    "modelrunner": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.modelrunner.run/mcp"]
    }
  }
}

Claude Code

claude mcp add --transport http modelrunner https://mcp.modelrunner.run/mcp

Then ask your assistant, for example: “Run meituan-longcat/longcat-video-t2v-480p on ModelRunner to generate video”. MCP setup guide.

Model Details

**LongCat-Video (Text-to-Video) turns plain language into polished motion.** Describe the scene you want—subjects, action, mood, and camera—and it generates a smooth, on-brand clip that looks like it was shot on set, not assembled from slides.

## How it works (for creators, not technicians)

You write a prompt in natural language. For example:

* “Product hero on a marble table; slow push-in; soft studio lighting; reflections on polished metal; minimal, luxury mood.” * “Cozy living room; morning light through windows; a cat jumps onto the sofa; gentle handheld camera; warm, cinematic tone.”

LongCat-Video translates that description into a coherent shot with consistent framing, lighting, and motion—ideal for product reels, teasers, explainers, and social posts.

## Directing the shot in plain language

* **Subject & action:** What’s in frame and what happens (“stainless steel watch rotating,” “barista pours latte art”). * **Camera feel:** Push-in, pull-back, orbit, dolly, crane, handheld, timelapse. * **Lighting & mood:** Soft studio, high-contrast, neon, natural daylight, golden hour. * **Style cues:** “Premium commercial,” “documentary,” “anime opening,” “stop-motion vibe.” * **Composition & scale:** Close-up, medium, wide; center-framed vs. rule-of-thirds.

Start simple, then iterate—each small prompt change nudges the look and motion.

## Timing & quality controls

* **`num_frames`** — Sets clip length. More frames = longer video. * **`fps`** — Playback speed (frames per second).

* **24–30 FPS** feels like live-action footage (great for products and demos). * **12–15 FPS** gives a stylized animation feel and smaller files. * **`num_inference_steps`** — Speed vs. polish.

* Fewer steps for fast drafts; more steps for cleaner finals. * **`seed`** — Repeatability. Use a seed to aim for similar motion and look across variations. Omit it for more variety.

## Output options

Choose how the file is delivered based on your workflow:

* **`video_output_type`** — Container/codec. * **`video_quality`** — Visual fidelity / bitrate level. * **`video_write_mode`** — Faster export vs. smaller file size.

These options let you create quick internal previews or high-quality clips ready for publishing.

## Prompt examples (copy/paste and adapt)

* “Matte-black earbuds on a linen fabric; slow top-down orbit, soft bounce light, subtle depth of field; premium commercial tone.” * “Urban rooftop at sunset; athlete ties shoes, camera dolly right then slight tilt up; crisp, energetic, motivational feel.” * “Cozy tabletop scene; pour-over coffee with visible steam; gentle push-in, natural morning light, warm color grade.”

## Common use cases

* **Product explainers & hero shots:** Script smooth, controlled motions that fit your brand. * **Brand bumpers & intros:** Generate short, memorable openers without a full production crew. * **Storyboard & previz:** Test timing, blocking, and mood before committing to a live shoot. * **Social ads & teasers:** Turn a short text brief into scroll-stopping vertical video. * **Mood films & style tests:** Explore lighting, camera language, and art direction quickly.

**Tip:** Keep prompts concrete (“slow push-in,” “soft top-light,” “center-framed close-up”) and iterate in small steps for the most predictable, on-brand results.

LongCat-Video t2v API

Model Input

Input

Model Output

Output

Model Example Requests

Examples

LongCat-Video t2v API

cURL

JavaScript

Python

Input parameters

Use LongCat-Video t2v from Claude & Cursor (MCP)

MCP client config (Claude Desktop, Cursor)

Claude Code

Model Details

Model Details