Seedance 2.0 Fast

ByteDancevideolive~2m 59s

Professional-grade video creation model with native audio, similar to SeeDance 2.0 but faster and cheaper.

avg$0.869/ run

An estimate, trued up after the run. Failed runs cost $0.00.

How this model is billed

Quoted before you run. A failed run costs $0.00.

Typical run ~2m 59s: the measured mean, end to end.

Compare every price

Run Seedance 2.0 Fast

Overview

Seedance 2.0 Fast, from ByteDance, is a professional-grade video model with native audio that is similar to Seedance 2.0 but faster and cheaper. It tops out at 720p, which is the trade: you give up the 1080p and 4K tiers and get quicker, lower-cost clips.

It is the one to reach for when you are iterating, producing social or product clips at volume, or want reference video support without paying for the larger model.

What it does well

Reference video. Attach up to 3 clips of 2 to 15 seconds each for motion or style transfer and cite them in the prompt as "video 1".
Reference images and audio. Up to 9 images, cited as "image 1", and up to 3 MP3 files, cited as "audio 1".
Native audio. Dialogue, effects, ambience and music generated with the picture.
Multi-shot prompts. "Shot 1:", "Shot 2:" for a sequence in a single clip.
Clip length. Choose any whole number of seconds from 4 to 15.

Prompting tips

Use 5 seconds for social posts and 10 to 15 for product demos. A first frame is best for product shots and character-led scenes, but it cannot be combined with reference images. Fewer references give each one stronger influence. If you supply audio, include an image or video input too: audio alone with text is not accepted.

Limits to know

Resolution is 480p or 720p. Requests that include reference videos are billed at the video-to-video rate. This model is billed on the real cost of each run, so the price depends on length and settings and is quoted on the Run button before you spend.

Parameters

What Seedance 2.0 Fast accepts, in the order the playground shows it. The full reference, with an example for each language, is on the API page.

Parameter	Type	Default	Allowed	What it does
`promptrequired`	string	—	—	Describe the video scene; use Shot 1: / Shot 2: for multi-shot or image 1 / video 1 for omni-reference
`first_frame_url`	string	—	one image url	Image URL to anchor the starting frame; mutually exclusive with reference_images
`last_frame_url`	string	—	one image url	Image URL to define the final frame; requires first_frame_url set
`reference_images`	string[]	—	up to 9 image urls	Up to 9 image URLs for style or character consistency; cite as image 1 in prompt
`reference_videos`	string[]	—	up to 3 video urls	Up to 3 reference videos (2-15 seconds each) for motion/style transfer
`reference_audios`	string[]	—	up to 3 audio urls	Up to 3 reference audio files (MP3)
`duration`	enum	5	4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15	Video length in seconds: 4, 5, 6, 8, 10, 12, or 15; -1 for auto
`resolution`	enum	720p	480p · 720p	Output resolution: 720p recommended for production; 480p for faster previews or low-bandwidth delivery
`aspect_ratio`	enum	16:9	16:9 · 9:16 · 1:1 · 4:3 · 3:4 · 21:9	Output dimensions: 16:9 for widescreen, 9:16 for mobile Reels, 1:1 for feed posts, 21:9 for cinematic
`generate_audio`	boolean	true	—	Enable native audio co-generation (dialogue, SFX, ambient, music); set true for final delivery, false for draft…
`return_last_frame`	boolean	false	—	Return the final frame as a separate image URL; useful for chaining clips or extracting scene-end stills
`bitrate_mode`	enum	standard	standard · high	Output video encoding bitrate
`auto_resize`	boolean	true	—	Automatically resizes an input image that falls outside the provider's accepted range (300-6000px per side, aspect…

Seedance 2.0 Fast pricing

Billed on the real cost of each run, which averages about $0.869 per run.

The exact price for your settings is on the Run button before you spend anything, and a failed run costs $0.00. Compare every model and plan.

Call Seedance 2.0 Fast from code

One POST returns a request id at once; poll it or take a webhook. Python, JavaScript, webhooks and every parameter are on the Seedance 2.0 Fast API page.

curlPOST /v1/models/seedance-2.0-fast

curl -X POST https://api.ezflows.io/v1/models/seedance-2.0-fast \
  -H "Authorization: Bearer $EZFLOWS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a wide shot of a desert at dusk",
    "duration": 5,
    "aspect_ratio": "16:9"
  }'

# 202 Accepted, at once — it never waits for the model:
# {"request_id": "req_8f2a41c9b70e4d15", "status": "queued",
#  "poll_url": "https://api.ezflows.io/v1/requests/req_8f2a41c9b70e4d15"}

Seedance 2.0 Fast: questions

How is this different from Seedance 2.0?

It is faster and cheaper, and limited to 480p and 720p. Seedance 2.0 goes up to 4K and does not take reference video.

Can I use a first frame and reference images together?

No. A first frame is mutually exclusive with reference images.

How much does Seedance 2.0 Fast cost?

Seedance 2.0 Fast is billed on the real cost of each run, which averages about $0.869 per run. The exact price for your settings is quoted before you run, and a failed run costs $0.00.

How long does a Seedance 2.0 Fast run take?

The measured average is ~2m 59s, end to end. Longer outputs and higher settings take longer. The API is asynchronous: you get a request id at once, then poll it or receive a webhook when the result is ready.

How do I call Seedance 2.0 Fast from code?

Send a POST to https://api.ezflows.io/v1/models/seedance-2.0-fast with your API key and the parameters as a JSON body. It answers 202 with a request id and a poll URL; poll it until the status is final, or register a webhook. Examples in curl, Python and JavaScript are on the API page.

Where are Seedance 2.0 Fast results stored, and for how long?

Results are saved to media.ezflows.io and kept for 90 days. Download anything you want to keep longer.