Seedance 2.0 Fast reference-to-video
doubao/seedance-2-0-fast/reference-to-video
Seedance 2.0 Fast (Video-Edit) edits an input video from a natural-language prompt at a faster, cheaper tier. Built on ByteDance Seed's unified multimodal architecture, it preserves subject identity, composition, and motion while rewriting lighting, style, weather, environment, or specific elements as instructed. Ready-to-use REST API, best performance, no coldstarts, affordable pricing.
Examples
Parameters
| Name | Type | Default | Constraints | Description |
|---|---|---|---|---|
| imagesReference Images | image_upload | — | image/* · 1–10 items | Optional reference image URLs to guide the edit (subject identity, style, etc.). |
| audiosReference Audios | audio_upload_group | — | audio/* · 0–10 items | Optional reference audio URLs to guide audio generation. |
| promptPrompt | textarea | — | ≤ 5000 chars | Describe the edit you want applied to the input video. The prefix "Edit the input video." is added automatically. |
| aspect_ratioAspect Ratio | select | — | 16:9 | 9:16 | 4:3 | 3:4 | 1:1 | 21:9 | Aspect ratio of the output video. Adapts to the input if not specified. |
| resolutionResolution | select | 720p | 480p | 720p | 1080p | Output video resolution. |
| duration *Duration | slider | — | 4 ~ 15 · step 1 | Output video length in seconds (4-15). Auto-detected from the input video if not specified. |
| generate_audioGenerate Audio | boolean | true | — | Whether to generate native audio for the edited output. Defaults to true. When set to false, the input video's audio track is preserved on the output instead. |
API
Call this model through one unified REST API. Get a key on the API Keys page.
cURL
# 1) Submit — returns { "task_uuid": "..." }
curl -X POST "https://www.namifusion.com/api/v1/marketplace/run/doubao/seedance-2-0-fast/reference-to-video" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": {
"images": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/e7daac690fea.jpeg",
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/e78f60fc4789.jpeg"
],
"prompt": "Transform the scene into a rainy cyberpunk night setting with glowing neon reflections on the pavement, cooler blue and magenta lighting, subtle fog in the background, and enhanced cinematic contrast while preserving the person'\''s identity, walking motion, camera movement, and overall composition.",
"aspect_ratio": "16:9",
"resolution": "720p",
"duration": 10,
"generate_audio": true
}
}'
# 2) Poll until status is "completed", then read the output URLs
curl "https://www.namifusion.com/api/v1/marketplace/run/tasks/TASK_UUID" \
-H "Authorization: Bearer YOUR_API_KEY"Python
import time, requests
API_KEY = "YOUR_API_KEY"
HEADERS = {"Authorization": f"Bearer {API_KEY}"}
# 1) Submit
resp = requests.post(
"https://www.namifusion.com/api/v1/marketplace/run/doubao/seedance-2-0-fast/reference-to-video",
headers=HEADERS,
json={
"input": {
"images": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/e7daac690fea.jpeg",
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/e78f60fc4789.jpeg"
],
"prompt": "Transform the scene into a rainy cyberpunk night setting with glowing neon reflections on the pavement, cooler blue and magenta lighting, subtle fog in the background, and enhanced cinematic contrast while preserving the person's identity, walking motion, camera movement, and overall composition.",
"aspect_ratio": "16:9",
"resolution": "720p",
"duration": 10,
"generate_audio": True
}
},
)
resp.raise_for_status() # 401/402/429/5xx stop here instead of polling a bad task
task = resp.json()
# 2) Poll until a terminal state (completed / failed / cancelled).
# This model is allowed up to 300s server-side.
deadline = time.time() + 360
while task.get("status") not in ("completed", "failed", "cancelled"):
if time.time() > deadline:
raise TimeoutError(f"still {task.get('status')} — keep the task_uuid and poll later")
time.sleep(3)
poll = requests.get(f"https://www.namifusion.com/api/v1/marketplace/run/tasks/{task['task_uuid']}", headers=HEADERS)
poll.raise_for_status()
task = poll.json()
print(task["status"], task.get("output"))JavaScript
const API_KEY = "YOUR_API_KEY";
const HEADERS = { Authorization: `Bearer ${API_KEY}` };
// 1) Submit
const resp = await fetch("https://www.namifusion.com/api/v1/marketplace/run/doubao/seedance-2-0-fast/reference-to-video", {
method: "POST",
headers: { ...HEADERS, "Content-Type": "application/json" },
body: JSON.stringify({
"input": {
"images": [
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/e7daac690fea.jpeg",
"https://assets-public.namifusion.com/marketplace/images/2026-06-12/e78f60fc4789.jpeg"
],
"prompt": "Transform the scene into a rainy cyberpunk night setting with glowing neon reflections on the pavement, cooler blue and magenta lighting, subtle fog in the background, and enhanced cinematic contrast while preserving the person's identity, walking motion, camera movement, and overall composition.",
"aspect_ratio": "16:9",
"resolution": "720p",
"duration": 10,
"generate_audio": true
}
}),
});
if (!resp.ok) throw new Error(`submit failed: ${resp.status} ${await resp.text()}`);
let task = await resp.json();
// 2) Poll until a terminal state (completed / failed / cancelled).
// This model is allowed up to 300s server-side.
const deadline = Date.now() + 360 * 1000;
while (!["completed", "failed", "cancelled"].includes(task.status)) {
if (Date.now() > deadline) throw new Error(`still ${task.status} — keep the task_uuid and poll later`);
await new Promise((r) => setTimeout(r, 3000));
const poll = await fetch(`https://www.namifusion.com/api/v1/marketplace/run/tasks/${task.task_uuid}`, { headers: HEADERS });
if (!poll.ok) throw new Error(`poll failed: ${poll.status}`);
task = await poll.json();
}
console.log(task.status, task.output);Documentation
Seedance 2.0 Fast Video Edit
Fast, affordable prompt-based video editing with strong subject, composition, and motion preservation.
Seedance 2.0 Fast Video Edit is ByteDance Seed’s cost-efficient video editing model for transforming existing videos with natural-language instructions. Built on a unified multimodal architecture, it can rewrite lighting, style, weather, environments, and specific scene elements while preserving subject identity, composition, and motion from the source footage.
🚀 Key Features
- Natural-language video editing: Describe the desired change in plain English, and the model applies the edit without requiring manual compositing workflows.
- Strong subject and motion preservation: Maintains faces, objects, camera movement, and overall scene continuity through the edit.
- Multi-reference guidance: Supports optional reference images and reference audio to steer identity, style, or soundtrack direction.
- Native synchronized audio: Can generate synchronized audio in the same pass, reducing post-processing overhead.
- Faster, lower-cost tier: Designed for teams that need scalable video editing throughput at a more affordable price point than the standard tier.
- No cold start: Well suited for production environments that value predictable responsiveness and operational efficiency.
🛠️ Technical Specifications
| Item | Details |
|---|---|
| Model architecture | ByteDance Seed unified multimodal video editing architecture |
| Model type | Video-to-video editing with natural-language control |
| Input | Video URL, Prompt, optional reference images, optional reference audio |
| Output | Edited Video |
| Output Format | Video(with optional native generated audio) |
| Resolution | 480p, 720p(default), 1080p |
| Duration | 4–15 seconds; auto-detected from input if not specified |
| Input video limit | Videos longer than 15 seconds are trimmed to the first 15 seconds |
| Aspect Ratio | 16:9, 9:16, 4:3, 3:4, 1:1, 21:9; adapts to input if unspecified |
| Audio behavior | Native audio generation enabled by default; if disabled, the original input audio track is preserved |
| Real-time context | Optional web search support |
| Latency profile | Fast tier optimized for lower latency and cost efficiency |
| Core strengths | Identity preservation, motion consistency, scene rewriting, no cold start |
Sample Prompts
- Change the scene to a rainy neon city at dusk, keep the subject’s motion and camera movement unchanged, and add a cinematic lighting style.
- Preserve the original composition and character identity, but transform the environment into a snowy winter street with realistic textures.
- Restyle the video as a premium fashion commercial with richer contrast, refined lighting, and polished material details while keeping facial features and movement intact.
💰 Pricing
| Service Type | Price | Estimated Cost |
|---|---|---|
| Video Gen (T2V/I2V) | $5.6 / 1M Tokens | ~$0.1 per second |
| Video Gen (V2V) | $3.3 / 1M Tokens | ~$0.065 per second |
💡 Best Use Cases
- E-commerce and brand marketing: Rework product videos into multiple seasonal, stylistic, or campaign-specific variants for ads and creative testing.
- Social media content production: Generate multiple edited versions of short-form videos quickly and cost-effectively.
- Previsualization and creative development: Explore alternate weather, lighting, and environment concepts while preserving original camera motion.
- Localization and campaign adaptation: Update scenes, atmosphere, and audio treatment for regional markets, events, or promotional themes.
🔗 Related Models
- Seedance 2.0 Video-Edit: Standard-tier version for users who may prefer the non-fast edition.
- Seedance 2.0 Fast Image-to-Video: Fast image-to-video generation for turning still images into motion content.
- Seedance 2.0 Fast Text-to-Video: Fast text-to-video generation directly from Prompt input.
- Bytedance Seedance 2.0 Fast Video Edit Turbo: A related variant for workflows prioritizing even faster turnaround.
Related models
xAI Grok Imagine Video v1.5 Reference to Video
Generate a 1-15 second video from a prompt and 1-7 reference images at 480p or 720p.
MiniMax H3 Reference to Video
MiniMax H3 Reference to Video generates coherent 2K videos from natural-language prompts and multimodal references, including images, videos, and audio, guiding subject consistency, motion, timing, visual style, and scene continuity. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Gemini Omni Flash Reference to Video API
Gemini Omni Flash Reference to Video creates short AI videos with synchronized audio from one or more reference images and a text prompt, preserving visual identity and following the provided references for guided multimodal video generation. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Seedance 2.0 Mini Reference to Video
Seedance 2.0 Mini Reference-to-Video is a high-speed, cost-effective video generation model developed by ByteDance, engineered specifically for creating multi-shot videos with a cinematic feel. By seamlessly combining reference images with text prompts, it generates highly narrative video segments. The model features advanced AI camera control and ensures robust character consistency across different scenes. It supports multiple output resolutions from 480P up to 4K, durations ranging from 4 to 15 seconds, and flexible aspect ratios. Complete with an out-of-the-box REST inference API, it delivers exceptional performance with zero cold starts, offering an affordable yet premium solution for professional video creation.
Seedance 2.5 Reference to Video
The "Reference-to-Video" feature of Seedance 2.5 is the ultimate solution for visual stylistic unity. It precisely extracts artistic styles, lighting tones, or compositional intents from reference materials and seamlessly integrates them into newly generated videos, ensuring a highly consistent visual language for your creative series.