How to AI Animate. AnimateDiff in ComfyUI Tutorial.

Text2Video and Video2Video AI Animations in this AnimateDiff Tutorial for ComfyUI.

CHAPTERS
0:00 AI Animate with AnimateDiff
0:42 Download workflows
1:18 How to use AnimateDiff - Text2Video, Settings Guide & Easy Cloud Solution
11:05 Video2Video + Local Install
22:44 Prompt travel/scheduling
25:48 How to install ffmpeg

[cosmiccarp]

Sweet, I literally just sat down to learn comfy and animatediff. Your timing is impeccable sir! Comfy UI is worth it everyone! I'm really liking it. Especially the screen shot workflows!

[ElHongoVerde]

Friday and new video from Seb. Nothing could go wrong today!

[PushedToInsanity]

Great tutorial as usual. Appreciate all the examples you've give, by far my favorite YouTuber when it comes to AI art. Do you offer personalized lessons?

[clovernacknime]

7:00 I think you mean convergence. Divergence means that a sampler (function) never settles down to (approaches) a particular image (or other output), while convergence means it will.

[defy_norms]

My gosh, dude you got a subscriber!! You are amazing, keep it up. This helps small Youtube channels like us immensely

[IAmTheAntiArt]

My first T2V SDXL AnimateDiff with HotShotXL vid is rendering right now! This vid got me started with animate diff in comfy 6 days ago. Thanks!

[digidope]

Context window: That is the fps that the model is trained for and you should NOT change it with animateDiff. If you use other model than AnimateDiff like HotShot then HS is trained with 12fps and then you need to change that value.

[KDawg]

Very nice tutorial! Just a heads up to those who want to use SDXL models; you can use them, but you have to use the Hot Shot motion module, and set the resolution smaller.

[chipko]

This is fantastic.. well produced, well explained and so informative. Thank you.

[KINGLIFERISM]

What you did that many do not... was you made errors and corrected them. That is a sign of a person that understands user experience. That is also a person of empathy. You sir... are a nice person.

[dustinjohnsenmedia]

Great video! Now I just figure out how to tack on an upscaler at the end

[choboruin]

First guide that actually let me generate something. Wasted 8 hours today using this and I just got a video after another 30 minutes and its just a puddle LOL. Appreciate the tutorial though! Great content.

[sharpenednoodles]

Finally, an animation tutorial from the king

[joonienyc]

bro, the best way to experiment with ur basic txt2vid and vid2vid ! thank u

[ShoreAllan]

Hello Sebastian,

many greetings from Berlin and thank you for your work! What do I have to do to extend your workflow with a face swap from a photo - i.e. video-and-picture-to-video!

[gorge.p]

Hey Sebastian. Nice video. Have you planned to make an AnimateDiff video for A1111?

[Shingo_AI_Art]

That thing about divergency with samplers was interesting, it would be nice to have a video reviewing all of them

[Mowgi]

Mate, you have great timing. I've just recently installed ComfyUI and started getting into animatediff, but I really haven't understood consistent animations, just the morphin stuff so far.

[Enchanted.Explorations]

Thank you so much! Very cool stuff! (Y)

[Tedrer]

Hi. There is some changes, animatediff combine is deprecated. it is the same to use video combine node?