WAN2.2 T2V for 8GB-VRAMlets!
Details
Download Files
Model description
🎬 Create an 8-second, 480x480px video from text input – in just ≈5 minutes on an average Gaming PC with 8GB VRAM & 32 GB RAM! 🎮⏱️
Welcome to the fast lane of video generation! You're about to make a short video in just ≈5 minutes on an average Gaming PC... which means you have just enough VRAM to generate a video clip, but definitely not enough to run a small town.
You are going to utilize the WAN2.2-14B-Rapid-AllInOne-GGUF model: a fast, optimized solution for Text-to-Video (T2V) and Image-to-Video (I2V) in a single model. It's built on the powerful WAN 2.2 14B model and optimized for speed and efficiency (FP8 precision) on consumer-grade hardware using the lightweight GGUF format.
💡 More LoRas and higher resolutions demand more (V)RAM - getting memory allocation errors? Simply lower the number of frames (the "length" value in the resolution node)!
What You Need
AllInOne GGUF Models Hugging Face - WAN2.2-14B-Rapid-AllInOne-GGUF
--> Examples were made with THIS ONE
Encoder GGUF Hugging Face - umt5-xxl-encoder-gguf
