Introducing Seedance 2.5, our new-generation video creation model! With advancements in long-form storytelling, multimodal reference, and editing, Seedance 2.5 goes beyond longer single-pass video generation, delivering the journey from idea to finished video with greater control. Key highlights include: 1. Users can generate high-quality, 30-second audio-video clips in a... single pass, bringing a complete story to life in one take. 2. Users can input up to 30 images, 10 video clips, and 10 audio clips as reference materials in a single pass. 3. Seedance 2.5 offers timestamp-level control for targeted editing of audio and video content, notably improving efficiency and controllability. Visit project homepage learn more：show more

it's gonna be a cook addition in our video generator 😎

a little more work with voice out of mouth needed...not quite natural.

yeayyy the new seedance 2.5

30s one take multimodal refs 🤯 Massive leap

Expect to see AI short films flooding X in the next few weeks

@capcutapp Much better for dialogue and pacing 👍

I'd like to check out the prompt - how it was done...

Timestamp level editing is the feature that actually matters here 🌊

Good

The site only has 2.0

Impressive leap forward making AI video creation more controllable

30 images and 10 videos?

👍

too bad there is no way for anyone to test it

30 second single pass video is wild, the multi reference input sounds super useful too

Advancing long form storytelling is the harder problem to solve.

What about ip characters? Wasnt there a new ip system to access them and use them in videos?

Amazing From idea to finished video. One pass, total control.

iLands creates a setting where economic outcomes become part of the learning process.

👏👏👏👏

looks good, getting close to imagine 1.5

In fairness I'm still waiting for somebody to beat this record on a secret workflow. I do use Seedance 2.0 and love it.

excited to see what creators build with this

I'm Nomi AI, an AI Researcher and Content Creator focused on AI, automation, and emerging technologies. I help innovative projects increase visibility and reach a targeted AI audience through strategic promotion. If you're looking to grow awareness for your project, I'd be happy to discuss a collaboration. Best regards, Nomi AI

Nice! I can now generate 30sec vid on one go😍

🙈考古，网红打卡

30 images, 10 videos, and 10 audio references? That’s serious creative control. 🚀

Excited to see more

The jump from generation to direction is the exciting part.

Seedance's motion consistency is impressive too, similar space to what we've been exploring with Wan3.0. Just posted a rain-confession clip with a long single take, worth a look:

The editing controls sound really useful.

This is a huge leap for AI video creation. The combination of 30-second single-pass storytelling, multimodal references, and timestamp level editing gives creators much more control from idea to final cut. Seedance 2.5 looks seriously impressive!

just wow

Nice share