Seedance 2.5
Continuous 30-second scenes with synchronized audio in one pass.
Professional video for every use case
Cinematic short scenes.
Build uninterrupted narrative beats with stable characters, environments, and pacing from opening shot to final moment.Multilingual dialogue spots.
Generate spoken performances with lip movement, ambience, and sound effects aligned to on-screen action.Social campaign videos.
Combine brand images, scripts, style frames, and audio references into vertical or widescreen campaign clips.Music-driven montages.
Shape cuts, movement, and visual rhythm from audio and video references inside one coherent sequence.Continuous 30-second scenes
Seedance 2.5 can generate a complete 30-second sequence without assembling separate short clips. It is designed to preserve characters, lighting, environments, and motion across longer scenes.
- •Up to 30 seconds in standard mode
- •Clean 4K video output
- •Stable subjects and environments
- •Multi-shot pacing in one generation
- •Extended-duration beta on supported services
Up to 50 multimodal references
Guide a generation with text, images, video clips, audio, scripts, and style material. Multiple references help define identity, appearance, movement, location, sound, and overall art direction.
- •Text, image, video, and audio inputs
- •Up to 50 combined references
- •Product and character guidance
- •Wardrobe and location references
- •Script, music, and camera cues
Synchronized sound and dialogue
Seedance 2.5 generates sound alongside the visuals so timing follows the action on screen. It can coordinate spoken performances, lip movement, ambience, music, and effects within the same scene.
- •Dialogue and ambient sound
- •Music and sound effects
- •Timing matched to visible action
- •Lip-synced spoken performances
- •Multilingual creative workflows
Reference-guided motion and local edits
R2V references provide more direct control over movement, spatial layout, and character interactions. Local editing can change selected objects or regions while preserving the rest of the video's composition and timeline.
- •Structured motion-path guidance
- •Green-screen motion references
- •Multi-character spatial control
- •Region-specific object replacement
- •Preserved lighting and continuity
How it works
1
Describe and reference your scene
Write the subject, action, setting, camera movement, lighting, and sound you want. Add images, clips, or audio when you need tighter control over identity, motion, style, or rhythm.
2
Choose framing and duration
Select the aspect ratio that fits your destination, such as widescreen or vertical. Set the available resolution and duration, using the longer setting when your scene needs a complete narrative arc.
3
Generate, review, and download
Generate the video and review motion, continuity, subject accuracy, and audio timing. Refine the prompt or references if needed, then download the finished clip.
Pricing for Seedance 2.5
Runs on credits — no per-model surcharges, no surprise billing.
330credits
per second of video
≈ 1650 credits for a 5-second clip
Show pricing details▾
ResolutionCredits
- 720pdefault330/ sec
- 1080p750/ sec
- 480p150/ sec
Credits work across every plan. See /pricing for credit packages.
Use SeeDance 2.5 via the API
SeeDance 2.5 is available through the BudgetPixel developer API — the same model the studio runs, supporting text-to-video, image-to-video, and reference-to-video. Pricing is metered in credits (150 credits/second at 480p, 330 credits/second at 720p, 750 credits/second at 1080p), charged only on success, with an API key available on Premium plans and above.
curl -X POST https://api.budgetpixel.com/v1/videos/seedance-2.5 \
-H "Authorization: Bearer $BUDGETPIXEL_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt": "your prompt here"}'No code? Use SeeDance 2.5 from Claude or Cursor
SeeDance 2.5 is also available through the BudgetPixel MCP server. Connect your account to Claude, Cursor, or any MCP-enabled agent and generate with SeeDance 2.5 by asking in plain language — same credits, same pricing, no code required.
Frequently asked questions
What is Seedance 2.5?▾
Seedance 2.5 is ByteDance's multimodal AI video model for generating longer, controlled scenes from text and reference assets. Its main capabilities include 30-second standard generation, 4K output, synchronized sound, multimodal references, and localized video refinement.
How does Seedance 2.5 work?▾
You provide a scene description and can add images, videos, audio, scripts, or style references. The model interprets these inputs together to generate the subjects, movement, camera direction, environment, and sound as a coherent video sequence.
What inputs does Seedance 2.5 accept?▾
Seedance 2.5 supports text-to-video and image-to-video workflows, along with image, video, and audio references. Its multimodal reference workflow can combine up to 50 inputs to guide characters, products, locations, motion, style, and sound.
What resolutions, lengths, and formats does Seedance 2.5 support?▾
Seedance 2.5 is promoted for 4K output and clips up to 30 seconds in standard generation. Some provider interfaces also list an extended beta mode reaching 180 seconds. Aspect ratios, frame rates, and downloadable file formats depend on the access point's current controls.
Does Seedance 2.5 generate audio?▾
Yes. Seedance 2.5 can generate dialogue, ambience, music, and sound effects with timing aligned to the visuals, and it can also use audio references for creative guidance. Spoken performances can include synchronized lip movement.
What can the 50 references control?▾
References can establish faces, products, wardrobe, locations, color palettes, camera behavior, movement, music, and other creative details. Assigning a clear purpose to each reference helps the model follow a complex production brief without relying on text alone.
How much does Seedance 2.5 cost?▾
Generation uses credits, with cost generally scaling according to video length, resolution, and selected features. Longer or higher-resolution clips usually require more credits than short drafts, so current pricing is shown before generation.
How does Seedance 2.5 compare to other AI video generators?▾
Seedance 2.5 is positioned around longer single-pass scenes, synchronized audio, extensive multimodal references, and localized editing. It is especially relevant when continuity and detailed creative direction matter more than generating an isolated short shot.
Can I use Seedance 2.5 videos commercially?▾
Seedance 2.5 can support commercial production workflows, but usage rights are governed by the terms of the service providing access to the model. Review those terms and confirm that you hold the necessary rights to any uploaded images, footage, audio, brands, or characters.
How do I write a good Seedance 2.5 prompt?▾
Describe the subject, setting, action, camera movement, lighting, visual style, and sound in a clear sequence. For longer scenes, specify what should happen at each stage and use references to lock details that should remain consistent.
Does SeeDance 2.5 have an API?▾
Yes — SeeDance 2.5 is available through the BudgetPixel developer API via the `POST /v1/videos/seedance-2.5` endpoint, supporting text-to-video, image-to-video, and reference-to-video. Generate an API key from the developer console (available on Premium plans and above) and see the full reference at docs.budgetpixel.com.
How much does the SeeDance 2.5 API cost?▾
API usage is billed in credits from the same balance your plan includes, at the standard metered rate: 150 credits/second at 480p, 330 credits/second at 720p, 750 credits/second at 1080p. Failed generations are never charged. You can estimate any request's exact cost with POST /v1/cost before running it.
Can I use SeeDance 2.5 with Claude or other AI agents (MCP)?▾
Yes — SeeDance 2.5 is exposed through the BudgetPixel MCP server, so Claude, Cursor, and any MCP-compatible agent can generate with it from a plain-language request. Connect your account at budgetpixel.com/mcp; generations use the same credit balance and pricing as the web studio and the API.