Image2Video. Stable Video Diffusion Tutorial.

SVD Tutorial in ComfyUI. How to easily create video from an image through image2video.

[MauiMauricio]

I can't wait for when you can prompt for the motion itself

[pedroserapio]

After paying and playing with Runway, I feel this workflow with ComfyUI output being more clean.

[Ekopop]

your jokes feel like AI too, and I call that dedication ^^

[aguyandhisguitars]

need some help, im getting Error occurred when executing KSampler: input must be 4-dimensional, when trying to do the stable difusion animate in comfy ui, i have an AMD 7900XT.

[michail_]

I like the "decode" model better, moreover with this model it generates not 25 but all 45 and 60 frames better. You can also add IPAdapter, AnimateDiff and whatever you want.

[midgardian]

I just finished the Windows installation, and it runs like molasses with no way to input text it just randomly makes a video from my image. So, I guess I have to go learn ComfyUI now, it just looks painful to use compared to A1111...

[Nutronic]

I see that it is supported in SD.Next, can you do a tutorial on how to install it to use with that?

[Julia-phej]

I want to use this without UI, is that possible? Can't seem to find any code for it. Just loading diffusers from Huggingface to Colab. Anyone? :)

[RoguishlyHandsome]

I find that it just make characters deform into melted mutants just after a few frames. I am puzzled that this was deemed worthy of release.

[jameshughes]

I spent a while trying to set this up and use it, before I ever realized that it's completely uncontrollable and slow. No prompting, no control.. you get what ever it spits out, it takes forever, and lots of times all it does is pan. It is very, very cool but I feel like it's still more of a toy than a tool. But the temporal coherence makes me drool. I really hope we get some control in the future.

[Rasukix]

will be good when on a111, comfy too much of a workflow explosion

[Nibot]

Is there a way to add a prompt node to tell it a camera direction?

[pedroserapio]

Let's do an AI Dads jokes using ComfyUI workflow!

[Jammyup]

Any word on this becoming available for A1111? I setup comfy UI and personally just hate the interface. I'd rather have less control and rip my hair out less often lol.

[evgenika]

Hello Sebastian, thank you for the video, is it possible to install ComfyUI with SVD for cloud services such as Collab or Runpod? Thank you for your response ahead.

[padn]

Hello, Sebastian. Noted, that you are using a lot of ComfyUI lately. So you are not using automatic1111 no more?
What Stabl Diffusion UI is the best right now in your opinion? Sorry for asking

[TheFutureThinker]

It would be good if there's a custom node like Runway Gen-2 Motion Brush. Then we can control motion from Img2vid.

[rudyruderabbit]

why cant we use this in fooocus ui? i like that the best!

[kannanjsk]

These examples wont work in mac due to conv 3d issues

[jonathaningram]

No way to add a prompt, like make the character smile or something ? Just having parallax isn't really useful.