Preview for WAN 2.1 Text-to-Video 1.3B (480P)

Run this workflow on InstaSD

Get started in minutes! Run this ComfyUI workflow online - no setup required.

Description

This workflow leverages WAN 2.1โ€™s Text-to-Video 1.3B model to generate high-quality 480P videos from text prompts. It integrates advanced video generation techniques with ComfyUIโ€™s flexible node-based workflow to create smooth and coherent animations from AI-generated frames.

๐ŸŽฏ Features

  • Text-to-Video Generation โ€“ Converts text descriptions into dynamic 480P videos.
  • Efficient VRAM Usage โ€“ Runs on consumer-grade GPUs with only 8.19GB VRAM required.
  • Temporal Consistency โ€“ WAN-VAE ensures stable motion and clear details across frames.

๐Ÿ’ก Use Cases

  • ๐ŸŽฌ Creative Content โ€“ Generate short animations for storytelling, marketing, or social media.
  • ๐ŸŽฎ Game & Virtual Worlds โ€“ Prototype in-game cinematics and animated backgrounds.
  • ๐Ÿ“บ AI-Assisted Filmmaking โ€“ Experiment with AI-generated video concepts and scene planning.

Models

FileDestinationSource
wan_2.1_vae.safetensors/ComfyUI/models/vaeDownload
umt5_xxl_fp8_e4m3fn_scaled.safetensors/ComfyUI/models/text_encodersDownload
wan2.1_t2v_1.3B_bf16.safetensors/ComfyUI/models/diffusion_modelsDownload

Nodes

CLIPTextEncodeVAELoaderCLIPLoaderKSamplerVAEDecodeVHS_VideoCombineEmptyHunyuanLatentVideoUNETLoader