⚡WAN 2.1 Tutorial: Generate FREE AI Videos with ComfyUI (Text-to-Video & Image-to-Video)

Wan 2.1 is one of the most powerful open-source AI video models available for free, capable of both text-to-video and image-to-video generation. In this video, I’ll show you how to install it on your computer, optimize the settings for the best quality, and even get it running on low VRAM.

(use this link to get $5 credit for your 1st $10 deposit)

⚠️ Disclaimer: Some links in the description are affiliate links. If you make a purchase through them, I may earn a small commission at no extra cost to you.

💎 Support me on Patreon:

🎵 Where I get my Music:

🔈 Join my Discord server:

🕒 Chapters:
0:00 WAN 2.1: Alibaba's Video Model
00:29 Installation Guide
01:16 Text to video
6:12 How to Prompt for Wan 2.1
6:53 Wan 2.1 t2v on Low VRAM
7:45 Wan 2.1 on Cloud GPU
8:13 Where to find generated videos
8:23 Upscaling
9:08 Image to video
12:28 Wan 2.1 vs Minimax vs Luma's Ray2
12:42 i2v on Low VRAM
13:27 Upscaling with Topaz Video AI
13:42 JUST DO IT!

Join me!

#AI #ComfyUI #aivideo

Who am I?
-----------------------------------------
My name is Mohamed Mehrez and I create videos around visual effects and filmmaking techniques. I currently focus on making tutorials in the areas of digital art, visual effects, and incorporating AI in creative projects.

[MDMZ]

These Open source models are getting really GOOD! Which model should I make a video on next ?

[LicbeAIMedia]

If you're a small YouTuber - don't give up. You're on the right path. <3 Btw amazing tutorial bro, thanks for sharing this

[BitJager]

ty updating confy ui fix my problem with one module, gl

[bizbenoina]

I'd like to see more such videos on your channel reviewing free AI programs for generating content as most of AI generators online has it's pricing and not completely free, thanks again so much for your hard work👍

[bizbenoina]

thanks for making this video it's really a gold!

[KiWelten]

Making Triton+sageattn work on Windows would be a great tutorial for WAN too!

[tonyg]

excellent work! surprisingly I got this working on a GTX 1660 ti lol would you mind explaining how i can add Teacache and DiffSynth-Studio to Comfyui to try to improve the generation speed please

[MUDHD]

I noticed that unlike "text to video", with "image to video" it does not follow well or ignores completely when you describe the movement that the camera must do. maybe I'm doing something wrong! do you have any advice?

[Kevinstats]

Ouch, 4 second video. I need at least 10 seconds, hopefully they can make it smoother in the future to achieve this.

[MuhammadZubair-si]

Thanks for 5p subscribers 🎉 plz keep supporting

[ZeroCool]

Let's go to the point, for a 4070 TI SUPER 16 VRAM what model i need and how much time will take to do a 10 seconds video?

[nquartuccio]

which model should i download if i have a beefy computer that can run it

[Kangi_Yatapi]

I tried the same workflow, I have 24Vram. The actions are not that bad, it follows my prompts, but all my works are all pixeled and vibrating, why?

[andrelip]

Thanks for the video! expired URL for the workflow

[Devicesai]

Hello, please, how do I save the video? I don't speak English, I have already translated it, but I still can't save the video. Thank you in advance, all the best!

[Strassenpraktikant]

Bro can you explain why the prompt does not match at all to what i typed in? Bro i uploaded a picture of me and said transform me to a bird change the background to a rainy forest. Cinematic 4k and so on i really followed the prompt i created it with chatgpt but the result had nothing to do with what i wanted to get.

[joonienyc]

the SAVEWEBM node doesnt seem to be appear, how to get this missing node ?

[zoroashura]

workflow isn't available anymore, is been deleted, please fix it

[RestLer-]

why do my generations take so long? Image to video takes almost an hour. 3060 12 GB video memory

[lucariodabosse]

How can you use wan2.1 to do text to image?

EDIT: It's literally as simple as changing the "length" to 1 lmao