Skip to main content
NewSeedance 2.5 is live→
ModelRunner
bytedance avatar

Seedance 1.0 Pro API

bytedance/seedance-v1-pro

Seedance 1.0 generates 1080P videos with smooth motion, rich detail, and diverse styles, while the pro version adds multi-shot narrative and advanced instruction following for cinematic results.

A4000,RTX4000
Priced by resolution
OpenAPI

Model Input

Input

Text prompt for video generation

Input image for image-to-video generation

Min: 3 - Max: 12

Video duration in seconds

Video resolution

Video aspect ratio. Ignored if an image is used.

Whether to fix camera position

Random seed. Set for reproducible generation

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

Model Output

Output

Loading
Logs (1 lines)

Model Example Requests

Examples

Example output 1Example output 2Example output 3Example output 4Example output 5Example output 6Example output 7

Model Pricing

Pricing

Model pricing varies by the target resolution of your output video.

480p
$0.03
per second of output video
or around 33 seconds for $1
720p
$0.06
per second of output video
or around 17 seconds for $1
1080p
$0.15
per second of output video
or around 7 seconds for $1

Seedance 1.0 Pro API

Seedance 1.0 Pro is a image-to-video AI model by bytedance. On ModelRunner it runs through a REST API or via MCP from any AI assistant, at $0.15 per second of video.

POST https://queue.modelrunner.run/bytedance/seedance-v1-pro

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/bytedance/seedance-v1-pro \
  -H "Authorization: Key $MRUN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "seed": 0,
    "prompt": "A skier glides over fresh snow, joyously smiling while kicking up large clouds of snow as he turns. Accelerating grad…",
    "duration": 3,
    "resolution": "480p",
    "aspect_ratio": "16:9",
    "camera_fixed": 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/bytedance/seedance-v1-pro/requests/$REQUEST_ID/status" \
  -H "Authorization: Key $MRUN_API_KEY"
curl "https://queue.modelrunner.run/bytedance/seedance-v1-pro/requests/$REQUEST_ID" \
  -H "Authorization: Key $MRUN_API_KEY"

JavaScript

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

const result = await modelrunner.subscribe("bytedance/seedance-v1-pro", {
  input: {
    "seed": 0,
    "prompt": "A skier glides over fresh snow, joyously smiling while kicking up large clouds of snow as he turns. Accelerating grad…",
    "duration": 3,
    "resolution": "480p",
    "aspect_ratio": "16:9",
    "camera_fixed": 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/bytedance/seedance-v1-pro",
    headers=headers,
    json={
      "seed": 0,
      "prompt": "A skier glides over fresh snow, joyously smiling while kicking up large clouds of snow as he turns. Accelerating grad…",
      "duration": 3,
      "resolution": "480p",
      "aspect_ratio": "16:9",
      "camera_fixed": false
    },
).json()

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

Input parameters

NameTypeRequiredDescription
promptstringyesText prompt for video generation
imagestring (uri)noInput image for image-to-video generation
durationintegernoVideo duration in seconds Default: 5.
resolutionenumnoVideo resolution Default: "1080p".
aspect_ratioenumnoVideo aspect ratio. Ignored if an image is used. Default: "16:9".
camera_fixedbooleannoWhether to fix camera position Default: false.
seedintegernoRandom seed. Set for reproducible generation

Machine-readable: OpenAPI schema · llms.txt

Use Seedance 1.0 Pro from Claude & Cursor (MCP)

Point Claude Code, Claude Desktop, Cursor, or any MCP client at the ModelRunner MCP server and Seedance 1.0 Pro 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 bytedance/seedance-v1-pro.

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 bytedance/seedance-v1-pro on ModelRunner to generate video”. MCP setup guide.

Model Details

Model Details

**Seedance 1.0** is a series of video generation models developed by the ByteDance **Doubao Large Model Team**. The **Seedance 1.0 Pro** version represents the large-parameter, high-capability variant of the series.

Key highlights include:

- **Multi-shot narrative support** with seamless transitions between different scenes. - **Advanced semantic understanding**, enabling accurate interpretation of natural language prompts. - **Instruction-following capabilities**, producing results that closely align with prompt details. - Ability to generate **1080P high-definition videos** with smooth motion, rich details, and cinematic qualities. - Broad stylistic versatility ranging from photorealism to artistic styles such as cyberpunk, illustration, or textured media.

In short, Seedance 1.0 Pro is designed for creators who want to bring written descriptions and still images to life in motion, with professional-grade visual quality.

---

## Core Capabilities

### 🎥 Video Generation - **Text-to-Video (T2V):** Convert text descriptions directly into moving video sequences. - **Image-to-Video (I2V):** Animate static images, optionally enhanced with descriptive text prompts. - **Resolution:** Outputs at **1080p full HD** for production-ready clarity.

### ⚡ Motion and Dynamics - Handles **subtle movements** (e.g., facial expressions, gestures) and **large-scale dynamics** (e.g., action scenes, environment changes). - Ensures **physical realism** and stability in sequences, avoiding unnatural distortions. - Robust handling of **multi-agent interactions** and **complex action choreography**.

### 🎬 Multi-Shot Support - Native ability to generate **multi-shot sequences**, enabling cinematic narratives. - Maintains **subject, style, and atmosphere consistency** across multiple shots. - Smoothly transitions across **temporal and spatial scene changes**.

### 🎨 Style and Aesthetics - Wide style coverage: **photorealism, cyberpunk, illustration, felt texture**, and beyond. - Faithful adherence to stylistic prompts, ensuring creators’ intended mood and look are preserved. - Always aims for **cinematic visual richness**, with balanced lighting, textures, and details.

### 📝 Prompt Understanding - Accurately **parses natural language**, reducing ambiguity in interpretation. - Offers **stable camera movement control** and perspective adherence. - Handles **complex multi-layered scene descriptions** with strong adherence to intent.

---

## Technical Specifications

- **Model Version:** 1.0 Pro - **Output Resolution:** 1080p - **Input Types:** - Text prompts - Images (for I2V mode) - **Video Length:** Multi-shot capable (specific limits depend on prompt configuration)

---

## Model Performance

Benchmarks conducted on **SeedVideoBench-1.0** and independent evaluations show:

- **High prompt adherence**, even with detailed multi-part descriptions. - **Strong motion quality** for both human and environmental dynamics. - **Competitive aesthetic quality** against peer video models. - **Consistent visual fidelity** for image-to-video tasks.

---

## Example Use Cases

- **Creative Content:** Storyboarding, short films, music videos, or experimental art. - **Film Prototyping:** Visualizing scripts, animatics, and pre-visualization. - **Commercial Media:** Advertising clips, product showcases, branded content. - **Education & Documentary:** Visualizing complex concepts and narratives. - **Fantasy/Surrealism:** Generating dreamlike or imaginative video content.

---

## Known Limitations

- **Benchmark Bias:** Performance measured against standardized datasets; real-world quality may vary. - **Prompt Dependency:** Complex or vague prompts may reduce generation consistency. - **Multi-Shot Consistency:** Stronger results depend on clarity and specificity of scene descriptions.