Home Creators Posts Import Register Favorites Logout

Content

Video : https://youtu.be/zO52xUBJlP0

In this video, we explore two powerful AI video models, CausVid and MovieGen, both fine-tuned from WAN 2.1. CausVid specializes in fast auto-regressive video generation, ideal for quick, short clips, while MovieGen focuses on cinematic-quality video with enhanced realism and higher resolutions like 720p and 1080p. We test these models in ComfyUI, comparing their performance, speed, and output quality, and provide practical tips for running them on consumer-grade hardware. Whether you're an AI enthusiast, content creator, or developer, this video offers valuable insights into leveraging these tools for AI-generated video production.

Who is this for?

AI researchers and developersContent creators exploring video generationTech enthusiasts interested in diffusion modelsAnyone working with WAN 2.1 or ComfyUI

Why it matters:

AI video generation is evolving rapidly, and fine-tuned models like CausVid and MovieGen push the boundaries of speed and quality. Understanding their capabilities helps creators choose the right tool for their projects, whether for fast prototyping or high-end cinematic visuals.

CausVidhttps://causvid.github.io/

https://github.com/tianweiy/CausVid

https://huggingface.co/tianweiy/CausVidMoviiGen

https://huggingface.co/ZuluVision/MoviiGen1.1

https://huggingface.co/Kijai/WanVideo_comfy/tree/main

Safetensors for ComfyUI

MoviiGen1.1

FP16: https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan2_1-MoviiGen1_1_fp16.safetensors

FP8: https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan2_1-MoviiGen1_1_fp8_e4m3fn.safetensors

CausVid

https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan2_1-T2V-14B_CausVid_fp8_e4m3fn.safetensors

CausVid as Lora work with Wan2.1 T2V model

https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan21_CausVid_14B_T2V_lora_rank32.safetensors

https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan21_CausVid_bidirect2_T2V_1_3B_lora_rank32.safetensors

Files

Comments

No comments found for this post.