Seedance 2.5. Thirty seconds in a single take.
ByteDance's newest video model generates clips at any length from 4 to 30 seconds, in six formats from square to ultrawide, with native audio, at a price built for daily production.
- 4 to 30 second clips
- 6 aspect ratios
- Native audio
- First and last frame control
Seedance 2.5 removes the biggest constraint in AI video: clip length. Where most models stop at 8 or 15 seconds, Seedance 2.5 renders up to 30 seconds in one generation, and you pick the exact second, so a clip can match a narration track instead of the other way around. Six aspect ratios, native audio, and first and last frame keyframes round out a spec sheet aimed squarely at volume production.
It is the model our own faceless-channel pipelines render on: long single-take shots mean fewer stitches per Short, and the pricing means iterating five prompt variants is routine rather than a budget decision. Output tops out at 720p, so draft here and re-render hero assets on Veo 3.1 when a destination demands 4K.
Up to 30 second clips
Any exact duration from 4 to 30 seconds. Carry a whole Short or product beat in one take, no stitching, no seams.
Six formats
16:9, 4:3, 1:1, 3:4, 9:16, and 21:9 ultrawide. One model covers every placement in a campaign.
Native audio
Ambience, effects, and speech generated with the clip, so drafts arrive watchable, not silent.
Keyframe control
Pin the first and last frames with images to lock openers, reveals, and transitions into place.
What to make with Seedance 2.5
Real prompts, and what comes back. Every one runs in Treza chat or as a node on the canvas.
“A 23 second dolly through a neon-lit market at night, rain on the lens”
One continuous take timed to a 23 second narration.
“A hedgehog explores a sunlit garden, macro wildlife lens, 9:16”
A vertical short-form clip ready for Shorts or TikTok.
“Ultrawide 21:9 flyover of a coral reef, documentary style, 30 seconds”
A cinematic banner loop no fixed-step model can match.
“Square 1:1 loop of espresso pouring in slow motion with sound”
A feed-ready loop with generated audio.
“Start on this logo frame, end on the packed product shot (both attached)”
A clip that opens and lands exactly on your brand art.
“A 30 second unboxing beat: hands open a kraft box, product reveal at the end”
A full demo beat in one generation.
Seedance 2.5 is one node on a canvas
Chain it with language, image, and video models, then publish the whole pipeline as an API.
- Step 01
Draft the brief
A language model node expands a one-line idea into a detailed, on-brand prompt before it ever reaches the model.
- Step 02
Generate with Seedance 2.5
Run variants in parallel, compare against other models on the same prompt, and keep every result in run history.
- Step 03
Publish it as an API
The pipeline becomes a versioned endpoint. Your product calls it with the typed /invoke API or any OpenAI SDK.
Seedance 2.5, answered
How long can Seedance 2.5 videos be?
Up to 30 seconds in a single generation, selectable to the exact second from 4 to 30. That is the longest single-clip range in the Treza lineup, roughly four times Veo 3.1's longest clip and 50% longer than Sora 2 Pro's.
How is Seedance 2.5 different from Seedance 2.0?
Seedance 2.5 doubles the maximum clip length from 15 to 30 seconds at a lower price per clip. Seedance 2.0 keeps one advantage: reference images, so it remains the pick when a product or character has to stay consistent across generations. Many pipelines use both.
What resolution does Seedance 2.5 output?
480p or 720p. For phone-first vertical content that is plenty; for hero assets shown large, draft on Seedance 2.5 and re-render the winning prompt on Veo 3.1, which goes up to 4K. On Treza that swap is one dropdown on the node.
How much does Seedance 2.5 cost on Treza?
Treza uses prepaid credits with no subscription. Each clip is charged at Seedance 2.5's rate from your balance, and only successful generations are charged. Credit packs start at $5 and never expire.
Can I call Seedance 2.5 through an API?
Yes. Build a pipeline that uses Seedance 2.5 and publish it as a versioned HTTP endpoint. Call the typed /invoke API from your product, or point any OpenAI SDK at the OpenAI-compatible endpoint. If a better model ships later, swap it on the node without changing the API your product calls.
Your next prompt could be production.
Generate your first video, image, or draft today. Prepaid credits, no subscription.