Seedance 2.5 Flash Reference to Video
Guide Seedance 2.5 Flash with your own media: up to 10 reference images, 5 reference videos and 5 audio tracks per generation, cited in the prompt as @Image1, @Video1, @Audio1. Keep characters, products and style consistent at about twice the speed of Seedance 2.5.
34% less than the priciest provider (3 compared). Same model, same output.
Prices last checked November 22, 2026 at 01:00 CET: Seedance 2.5 Flash Reference to Video costs from $0.0706 per second, audio included on Scenetra, billed at provider cost with no markup. Each run is priced across 3 providers and runs on the cheapest one that supports your settings. Every price is re-checked against the providers daily at 01:00 CET.
Up to 10 reference images for character, style or scene. Cite in the prompt as @Image1 and so on.
Up to 5 reference videos (mp4/mov), each 2-30s, combined duration up to 30s. Cite as @Video1. Reference video seconds are billed as input.
Up to 5 reference audio files (wav/mp3), combined up to 30s.
Duration
Duration of the video in seconds (4-30). Billed per second.
Resolution
Video resolution.
Aspect Ratio
Aspect ratio of the generated video. 'adaptive' follows the prompt or the primary media.
Audio
Synchronized audio: voice, sound effects and music generated in the same pass.
34% less than the priciest provider (3 compared). Same model, same output.
Your result renders here.
Use Seedance 2.5 Flash Reference to Video via the API
Generate with Seedance 2.5 Flash Reference to Video from your own code. Async by design: submit a generation, get a run id back instantly, poll until it completes. Billed in credits at the best available provider price — from $0.0706 per second, audio included.
1 — Start a generation
curl -X POST https://app.scenetra.com/api/v1/generate/seedance-2-5-flash-ref2v \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"@Image1 walks through a sunlit market holding @Image2, handheld follow shot, crowd murmur","reference_images":["https://example.com/person.jpg","https://example.com/product.jpg"],"duration":"5"}'Response (202)
{
"runId": "run_abc123",
"status": "PENDING",
"model": "seedance-2-5-flash-ref2v",
"estimatedCredits": 793,
"statusUrl": "/api/v1/runs/run_abc123"
}2 — Poll until COMPLETED (every 2-5 seconds)
curl https://app.scenetra.com/api/v1/runs/run_abc123 \
-H "Authorization: Bearer YOUR_API_KEY"Response
{
"runId": "run_abc123",
"status": "COMPLETED",
"progress": { "completed": 1, "total": 1 },
"outputs": [
{ "type": "video", "url": "https://images.scenetra.com/..." }
],
"error": null,
"estimatedCredits": 793
}Request fields
| Field | Type | Description | Default |
|---|---|---|---|
| prompt* | string | Cite reference inputs in order with @-syntax: @Image1, @Video1, @Audio1. | — |
| reference_images | string[] (https URLs) | Up to 10 reference images for character, style or scene. Cite in the prompt as @Image1 and so on. | — |
| reference_videos | string[] (https URLs) | Up to 5 reference videos (mp4/mov), each 2-30s, combined duration up to 30s. Cite as @Video1. Reference video seconds are billed as input. | — |
| reference_audios | string[] (https URLs) | Up to 5 reference audio files (wav/mp3), combined up to 30s. | — |
| duration | string | Duration of the video in seconds (4-30). Billed per second.4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 | 5 |
| resolution | string | Video resolution.480p · 720p | 720p |
| aspect_ratio | string | Aspect ratio of the generated video. 'adaptive' follows the prompt or the primary media.adaptive · 21:9 · 16:9 · 4:3 · 1:1 · 3:4 · 9:16 | adaptive |
| generate_audio | string | Synchronized audio: voice, sound effects and music generated in the same pass.false · true | true |
| callbackUrl | string (https URL) | Optional. When the run finishes (completed or failed), we POST the final run status to this URL — same shape as the poll response, plus model, params, and chargedCredits. Best-effort; polling remains the source of truth. | |
Unknown fields are rejected with a helpful 400. Media fields take https URLs. There is no provider parameter — every run is routed to the cheapest capable provider automatically.
Good to know
Billing
Runs bill your account credits at provider cost with 0% markup. The submit response includes the credit estimate; failed generations are refunded automatically. Add ?dryRun=1 to get the estimate without running anything.
Errors
Errors are always { "error": { "code", "message" } } with stable codes: unauthorized, model_not_available, invalid_parameter, insufficient_credits, rate_limited. A failed run is not an HTTP error — poll responses report status: "FAILED" with the reason.
What makes Seedance 2.5 Flash Reference to Video different
Near-flagship quality, about twice as fast
Seedance 2.5 Flash keeps what makes Seedance 2.5 stand out: 30-second single-pass generation, native audio, multimodal references and strong instruction following. ByteDance rates it the closest model yet to the 2.5 flagship and stronger overall than Seedance 2.0, while it renders about twice as fast.
Lower cost per second
Flash costs roughly a third less per second than Seedance 2.5 at the same resolution. That makes it the practical tier for drafts, A/B variants, ad tests and daily production runs: settle framing, motion and sound on Flash, then render the final on Seedance 2.5 if you need it.
Up to 30 seconds with synchronized audio
Pick any whole-second duration from 4 to 30. Dialogue, sound effects and music are generated together with the picture, lip-synced, with no separate audio pass. Switch audio off with one toggle when you only need silent footage.
One model, three modes
Text to Video is one of three Seedance 2.5 Flash modes on Scenetra. Animate a first frame (with an optional last frame) with Image to Video, or guide generations with up to 10 reference images, 5 videos and 5 audio tracks with Reference to Video, cited in the prompt as @Image1, @Video1, @Audio1.
Parameters
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt* | text | Text prompt describing the desired video. Supports English and Chinese. | — |
| Duration | select | Video duration from 4 to 30 seconds. | 5 |
| Resolution | select | Output video resolution.480p · 720p | 720p |
| Aspect Ratio | select | Video aspect ratio, or 'adaptive' to select automatically from prompt context.16:9 · 4:3 · 1:1 · 3:4 · 9:16 · 21:9 · adaptive | adaptive |
| Generate Audio | select | Synchronized audio — voice, sound effects, and background music — generated with the video.false · true | true |
Pricing
From $0.0706 per second, audio included
Pay per generation — only pay for what you use.
480p
$0.0706/s
per second
720p
$0.1586/s
per second
What a video costs
| Duration | 480p | 720p |
|---|---|---|
| 5 seconds | $0.353 | $0.793 |
| 10 seconds | $0.703 | $1.58 |
| 30 seconds | $2.102 | $4.724 |
Use cases
Film & storytelling
- Complete 30-second scenes
- Dialogue clips with lip-sync
- Establishing shots
- Previz and animatics
Marketing & ads
- Product spots with music
- UGC-style ads
- Brand stings
- Concept testing
Social content
- Vertical 9:16 stories
- 30-second narratives
- Sound-on hooks
- Series with consistent style
E-commerce
- Product showcase clips
- Lifestyle b-roll
- Seasonal campaign variants
- Listing videos
Frequently asked questions
How many reference files can I use?+
Up to 10 reference images, up to 5 reference videos (each 2-30 seconds, 30 seconds combined) and up to 5 audio tracks (30 seconds combined) per generation.
How do I reference my uploads in the prompt?+
Use @-syntax in upload order: @Image1 is your first image, @Video1 your first video, @Audio1 your first audio track. For example: 'The character from @Image1 walks through the street in @Image2, moving like @Video1.'
How are reference videos billed?+
Reference video seconds are billed as input alongside the generated seconds, at a lower per-second rate than plain generation. Reference images and audio add nothing to the price.
Can I keep a character consistent across videos?+
Yes. Reuse the same reference images in every generation and cite them in each prompt; the model keeps the identity stable from clip to clip.
Related models
Seedance 2.5
ByteDance's Seedance 2.5 generates cinematic videos up to 30 seconds from text prompts with synchronized audio — voice, sound effects, and background music — built in by default. The newest Seedance generation, with mp4 or high-fidelity mov output for multi-round editing pipelines.
View model →Seedance 2
ByteDance's Seedance 2.0 generates cinematic videos from text prompts with synchronized audio — voice, sound effects, and background music. Pick any duration from 4 to 15 seconds and resolutions up to 4K, with per-second pricing that keeps 480p drafting cheap.
View model →Kling 4.0
Kuaishou's Kling 4.0 generates cinematic 3-30 second videos with native audio at 720p or 1080p, in 16:9, 9:16, 1:1 or 21:9 widescreen. Animate a first frame through up to 8 keyframes into a last frame, or keep characters and objects consistent from reference images and clips.
View model →Start creating with Seedance 2.5 Flash Reference to Video
Use Seedance 2.5 Flash Reference to Video alongside 50+ other AI models in Scenetra's visual workflow editor. No setup required.
Get Started