Seedance 2.0 Fast

Accelerated variant of ByteDance’s Seedance 2.0 with faster generation and lower per-second pricing.

Overview

Seedance 2.0 Fast (doubao-seedance-2.0-fast) is the accelerated variant of ByteDance’s Seedance 2.0 model. It offers the same feature set with faster generation speeds and lower per-second pricing — ideal for rapid iteration and high-volume production.

Property	Value
Slug	seedance-2.0-fast
Type	Video
Billing	Per-second × duration
480p price	$0.1168/second
720p price	$0.2512/second
480p + video ref	$0.0696/second
720p + video ref	$0.1504/second
Durations	4–15 seconds
Resolutions	480p, 720p

---

Supported Parameters

model · string · required

Must be seedance-2.0-fast

prompt · string · required

Video content description. Clearly specify subject, action, camera movement, and style for best results.

duration · integer

Video duration in seconds. Range: 4 to 15. Default: 5.

aspect_ratio · string

Output aspect ratio. Options:

• 16:9 — Landscape (default)
• 9:16 — Portrait
• 1:1 — Square
• 4:3 — Traditional
• 3:4 — Vertical traditional
• 21:9 — Ultra-wide

resolution · string

Output resolution. Options:

• 480p — Standard definition (default, $0.1168/s)
• 720p — High definition ($0.2512/s)

image_urls · array

Array of publicly accessible image URLs for image-to-video generation. Maximum 9 images.

Cannot be used together with the first/last frame parameters in the same request.

generate_audio · boolean

Generate accompanying AI audio with the video. Default: false.

camera_fixed · boolean

Fix the camera position during generation. Default: false.

web_search · boolean

Enable web search to reference online information during generation. Default: false.

return_last_frame · boolean

Return the last frame image URL with the result, useful for chaining sequential video clips. Default: false.

seed · integer

Random seed for controlling generation randomness. Same seed with same prompt produces similar (not identical) results.

video_urls · array

Array of publicly accessible video URLs to use as reference input. Maximum 3 URLs.

When provided, the generation uses the reference video to guide output and receives discounted pricing:

• 480p: $0.0696/s (vs$0.1168/s) — ~40% savings
• 720p: $0.1504/s (vs$0.2512/s) — ~40% savings

Supported formats: MP4, WebM, MOV.

audio_urls · array

Array of publicly accessible audio URLs to use as reference input. Maximum 3 URLs.

Supported formats: MP3, WAV, M4A, OGG.

---

Pricing

Seedance 2.0 Fast uses per-second billing with resolution-based pricing:

Standard (text/image-to-video)

Resolution	Price per second	5s video	10s video	15s video
480p	$0.1168	$0.584	$1.168	$1.752
720p	$0.2512	$1.256	$2.512	$3.768

With Video Reference (video-to-video)

When you provide a reference video via video_urls, the price is automatically discounted:

Resolution	Price per second	5s video	10s video	15s video
480p	$0.0696	$0.348	$0.696	$1.044
720p	$0.1504	$0.752	$1.504	$2.256

Seedance 2.0 Fast is 20% cheaper than Seedance 2.0 at both resolutions, making it the best choice for batch production and rapid prototyping.

---

Examples

Text-to-Video (480p)

curl -X POST https://veogen.studio/api/v1/generations \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-fast",
    "prompt": "A hummingbird hovering near a tropical flower, macro shot",
    "aspect_ratio": "16:9",
    "duration": 5,
    "resolution": "480p"
  }'

curl -X POST https://veogen.studio/api/v1/generations \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-fast",
    "prompt": "City time-lapse transitioning from day to night with streaking lights",
    "aspect_ratio": "21:9",
    "duration": 12,
    "resolution": "720p",
    "generate_audio": true
  }'

curl -X POST https://veogen.studio/api/v1/generations \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-fast",
    "prompt": "Gentle zoom into the subject with soft bokeh in the background",
    "aspect_ratio": "9:16",
    "duration": 8,
    "resolution": "480p",
    "image_urls": ["https://example.com/portrait.jpg"]
  }'

curl -X POST https://veogen.studio/api/v1/generations \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "seedance-2.0-fast",
    "prompt": "Apply cinematic color grading and slow-motion effect",
    "duration": 8,
    "resolution": "480p",
    "video_urls": ["https://example.com/reference.mp4"]
  }'

import requests
import time

token = "YOUR_TOKEN"
base = "https://veogen.studio/api/v1"
headers = {"Authorization": f"Bearer {token}", "Content-Type": "application/json"}

# Generate a quick 5s 480p video
res = requests.post(f"{base}/generations", headers=headers, json={
    "model": "seedance-2.0-fast",
    "prompt": "A chef plating an elaborate dessert, tracking shot",
    "aspect_ratio": "16:9",
    "duration": 5,
    "resolution": "480p"
})
data = res.json()
gen_id = data["data"]["id"]
print(f"Submitted: {gen_id} — charged ${data['price_usd']}")

# Poll until complete
while True:
    res = requests.get(f"{base}/generations/{gen_id}", headers=headers)
    gen = res.json()["data"]
    if gen["status"] == "completed":
        print(f"Done! URL: {gen['output']['url']}")
        break
    elif gen["status"] == "failed":
        print(f"Failed: {gen['error']}")
        break
    time.sleep(3)

---

Comparison with Seedance 2.0

Feature	Seedance 2.0 Fast	Seedance 2.0
480p price	$0.1168/s	$0.14512/s
720p price	$0.2512/s	$0.31232/s
480p + video ref	$0.0696/s	$0.088/s
720p + video ref	$0.1504/s	$0.1888/s
Quality	High	Highest
Speed	Faster	Standard
Parameters	All supported	All supported

For maximum quality output, consider Seedance 2.0.