Video Model — text, image & multimodal-to-video
Seedance 2.5 Flash is the lighter tier of ByteDance's Seedance 2.5. It keeps the 2.5 feature set: 4–30 second clips in a single pass, first and last frame, multimodal references, editing and extension, with audio co-generated alongside the visuals.
What it gives up is headroom: output tops out at 720p, the reference stack is 20 inputs instead of 50, and there is no Draft Mode. Reach for it for iteration and social-resolution work; move to Seedance 2.5 when the shot needs 1080p or a heavy reference stack.
Variants at a Glance
| Variant | Type | Modalities | Max Res. | Talking Points |
|---|---|---|---|---|
| 2.5 Flash | Video gen / edit / extend | T2V, I2V, first & last frame, multimodal refs | 720p | Same modes as 2.5; up to 20 reference inputs (10 img / 5 vid / 5 aud); 480p or 720p; no Draft Mode |
| 2.5 | Video gen / edit / extend | T2V, I2V, first & last frame, multimodal refs | 1080p | Up to 50 reference inputs (30 img / 10 vid / 10 aud); 480p draft, then final render at 1080p; shown for comparison |
Key Features
Generation Modes
- Text-to-Video (T2V): Generate video from a text prompt
- Image-to-Video (I2V): Text prompt plus a first-frame image
- First & Last Frame: Text prompt plus first-frame and last-frame images; if the two aspect ratios differ, the first frame wins and the last frame is auto-cropped
- Native 30s in one pass: A full 30-second clip is generated in a single pass, with no stitching
- Multimodal reference generation: Feed image, video, and audio references and specify in the prompt what each is for: subject, background, style, camera movement, motion, dialogue, music, or SFX
Editing & Extension
- Video editing: Replace subjects, add / remove / modify elements, or scope a change to a time window, driven by reference materials plus the prompt
- Extension: Continue a clip forward from its last frame, generate the moment before it, or bridge two clips
Technical Capabilities
| Duration | Any integer 4–30s per clip, generated natively in a single pass |
|---|---|
| Quality / Resolution | 480p, 720p (default 720p). No 1080p |
| Aspect ratios | 16:9, 9:16, 4:3, 3:4, 1:1, 21:9. If a specified ratio differs from an uploaded image's, the image is center-cropped |
| Reference inputs | Up to 20 total = 10 images + 5 videos + 5 audio clips. Formats and per-file limits match Seedance 2.5: videos and audio 2–30s per clip, each type under 30s total |
| Audio | Native, co-generated with the visuals; audio references can drive dialogue, music, and SFX |
Credit Cost
| Credits | Charged per | |
|---|---|---|
| Text-to-Video 480p | 200 | second |
| Text-to-Video 720p | 300 | second |
| Image-to-Video 480p | 200 | second |
| Image-to-Video 720p | 300 | second |
| Reference-to-Video, audio references 480p | 200 | second |
| Reference-to-Video, audio references 720p | 300 | second |
| Reference-to-Video, image references 480p | 200 | second |
| Reference-to-Video, image references 720p | 300 | second |
| Reference-to-Video, video references 480p | 200 | second |
| Reference-to-Video, video references 720p | 335 | second |
| Video editing 480p | 200 | second of input video |
| Video editing 720p | 335 | second of input video |
Seedance 2.5 Flash is not included in any Unlimited package; every generation uses credits. Credit costs can change. For the current price, check the model in Artlist AI Tools.
Limitations
- Output is capped at 720p. Anything that needs 1080p goes to Seedance 2.5
- No Draft Mode, and a Seedance 2.5 draft can't be finalized on Flash
- Reference stack is 10 images, 5 videos and 5 audio clips, well under the 50 that 2.5 takes
- Reference video and reference audio are each capped at 30s total per request (each file 2–30s)
- Single clip is capped at 30s; long-form still needs to be assembled from multiple generations
Prompting Tips
- State clearly in the prompt which reference materials to use and for what (subject vs style vs motion vs audio); the model follows your assignments
- With 20 reference slots, pick the few references that matter and label each one; if a shot needs more than 10 images, it's a Seedance 2.5 job
- For first & last frame generation, match the aspect ratios of both images, otherwise the last frame is auto-cropped to fit the first
- For edits, give a timestamp range to scope the change and add "keep everything else unchanged" so the model doesn't touch unrelated areas
- Describe the audio you want in the prompt; dialogue, music, and SFX are co-generated with the visuals, not added in a separate pass