Stuck with those super short AI video clips? We all are. You get a great start, and then… it just ends. So frustrating.
Well, hold on, because I just saw this game-changing post from Tianyu Xu that shares a brilliant way to create AI videos of basically unlimited length. This is huge.
He points out that the secret is using a model that supports start and end frames, like the new Kling v2.1.
Here’s the simple-but-genius method Tianyu Xu laid out:
⛓️ The Chaining Technique
- First Clip: You generate a video using Image A as the start frame and Image B as the end frame.
🎬 {Frame A —> Frame B} - Second Clip: Then, you take Frame B (the last frame of your first video) and use it as the start frame for your next video, ending on Frame C.
🎬 {Frame B —> Frame C} - Rinse & Repeat: You just keep this chain going (C to D, D to E…) for as long as you want. Mind. Blown.
💡 Bonus Trick from Tianyu Xu: You can even create perfect loops by making your very last frame the same as your first one! {Frame A → B → C → A}
The Reality Check ✅
Tianyu Xu is quick to point out it’s not perfect yet. The speed and motion can be a little inconsistent between the clips you stitch together.
But honestly? As he says, our attention spans are shorter than a goldfish’s anyway. Long, continuous shots rarely work on social media.
His advice is golden: be flexible. Use this frame-locking trick, cut your videos often, and edit freely instead of hoping for one perfect, long clip.
This is such a clever way to work around current limitations. For all the details, go check out Tianyu Xu’s original post!