ComfyUI Tutorial Series Ep 36: WAN 2.1 Installation – Turn Text & Images into Video!

Sobre esta lección

In this episode, we go through the step-by-step process of setting up the WAN 2.1 model in ComfyUI, allowing you to create videos from both text and images—right on your own computer! 🚀 We’ll start by updating ComfyUI, installing the necessary nodes, and setting up the required models. You’ll also get a clear breakdown of how different versions of the WAN model impact video quality, speed, and compatibility with your hardware. Whether you're working with 1.3B, 14B, or GGUF models, this guide will help you find the best fit for your workflow. 🔧 Additionally, we explore frame rate settings, video length calculations, and how to fine-tune prompts for optimal results. By the end, you’ll have everything you need to generate smooth AI-driven animations with ComfyUI! 🎬 Don’t forget to grab the free workflows on Discord and join the conversation with other creators. Let’s bring your AI video ideas to life! ✨ Get the workflows and instructions from discord for free https://discord.gg/gggpkVgBf3 Check Other Episodes https://www.youtube.com/playlist?list=PL-pohOSaL8P9kLZP8tQ1K1QWdZEgwiBM0 --- Test Wan 2.1 Image to Video 14B 480p workflow on RunningHub, https://www.runninghub.ai/post/1896577103540682754?utm_source=kol02-RH003 use one of these settings for widht and height Landscape 16:9 → 832x480px Portrait 9:16 → 480x832px Square 1:1 → 640x640px Add a text prompt in the positive prompt node For 16 FPS: Duration → Length Value (in the Latent Video Node) 1 second → 17 frames 2 seconds → 33 frames 3 seconds → 49 frames 4 seconds → 65 frames 5 seconds → 81 frames --- Unlock exclusive perks by joining our channel: https://www.youtube.com/channel/UCmMbwA-s3GZDKVzGZ-kPwaQ/join -- Comfy Blog Post for Wan 2.1 https://blog.comfy.org/p/wan21-video-model-native-support --- Update ComfyUI Then Install this nodes from Manager if you don't have them already - ComfyUI-GGUF - ComfyUI-VideoHelperSuite --- Model Links are in the workflows, but here are some of the links Go to this page and choose a diffusion model: https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/tree/main/split_files/diffusion_models Place in ComfyUI/models/diffusion_models Or gguf i2v model for 720p https://huggingface.co/city96/Wan2.1-I2V-14B-720P-gguf/tree/main Or gguf i2v model for 480p https://huggingface.co/city96/Wan2.1-I2V-14B-480P-gguf/tree/main Info about the models: - 1.3B models are smaller and work on GPUs with less VRAM. - 14B models are much larger and need a high-VRAM GPU. Download this text encoder model https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors?download=true Place in ComfyUI/models/text_encoders Download this clip model https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/clip_vision/clip_vision_h.safetensors?download=true Place in ComfyUI/models/clip_vision Download this vae model https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/vae/wan_2.1_vae.safetensors?download=true Place in ComfyUI/models/vae Restart ComfyUI Load the workflows #comfyui #imagetovideo #texttovideoaifree