Hey everyone 👋 I’m an indie maker obsessed with AI video, and I kept running into the same frustration: most tools feel like slot machines. You type a prompt, hit generate, and hope something usable comes out. With Seedance 2.0 I wanted to try a different path—treat AI more like a camera crew you can actually direct. Seedance 2.0 lets you mix text, images, clips, and audio as references so you can control shots, motion, and rhythm instead of relying on one vague prompt. I’ve been using it to prototype storyboards, remix existing footage, and test ideas for short explainers and ads without opening a full NLE. This launch is the first public version of that experiment. I’d love feedback from creators, filmmakers, and other indie hackers: what feels missing, what’s confusing, and what kind of control you wish AI video tools gave you. Thanks for checking it out!

seedance2.page

Direct multimodal AI video with Seedance 2.0

Direct multimodal AI video with Seedance 2.0