A Wan 2.1 ATI workflow. ATI stands for any trajectory instruction. What it does is, it moves parts of the image in different directions, defined by paths. These paths can be adjusted in the workflow.
Runtime profile
Source description
Description
A Wan 2.1 ATI workflow. ATI stands for any trajectory instruction. What it does is, it moves parts of the image in different directions, defined by paths. These paths can be adjusted in the workflow.
Time
The example video was generated in 420 seconds.
Requirements
This workflow was generated with 16 gb vram. Minimum requirement is 12 gb vram. And you should not have fewer than 32 gb system ram.
Estimated VRAM requirement
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Use Wan 2.1 ATI (ATI means any trajectory instruction) to move parts of an image in different directions along adjustable paths; it takes an image and path definitions and produces a video.
Input: an image and paths that define movement. Output: a generated video.
Wan 2.1 ATI is listed as a workflow with SD 1.5 as its base model and Arunderan as its publisher.
The listed minimum is 12 GB of VRAM and at least 32 GB of system RAM; the example used 16 GB of VRAM.
The workflow requires these node packs: comfyui_fill-nodes, comfyui-frame-interpolation, comfyui-kjnodes, and comfyui-videohelpersuite.
The workflow requires these model files: umt5_xxl_fp8_e4m3fn_scaled.safetensors, wan_2.1_vae.safetensors, wan2.1_vace_14B_fp16.safetensors, Wan21_CausVid_14B_T2V_lora_rank32.safetensors, clip_vision_h.safetensors, and Wan2_1-I2V-ATI-14B_fp8_e4m3fn.safetensors.
Adjust the paths to define the directions in which image parts move.
Suggestion · not verified
Treat 420 seconds as an example-video figure, not a runtime estimate for your setup.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
1 excerptSource context: 69 downloads · Type Workflows · Base model SD 1.5
Estimate unavailable
304 MB across 1 of 6 model files. Model file total + 25% loading overhead + 2 GB execution buffer, rounded up.
Requirements
10 requirementsVision model · Unknown
umt5_xxl_fp8_e4m3fn_scaled.safetensors
File unverifiedText encoder · Hugging Face · Comfy-Org/Wan_2.1_ComfyUI_repackaged · Comfy-Org/Wan_2.1_ComfyUI_repackaged · Hugging Face
VAE · Hugging Face · Comfy-Org/Wan_2.1_ComfyUI_repackaged · Comfy-Org/Wan_2.1_ComfyUI_repackaged · Hugging Face
Wan2_1-I2V-ATI-14B_fp8_e4m3fn.safetensors
File unverifiedCheckpoint · SAFETENSORS · Unknown
wan2.1_vace_14B_fp16.safetensors
File unverifiedCheckpoint · Hugging Face · Comfy-Org/Wan_2.1_ComfyUI_repackaged · Comfy-Org/Wan_2.1_ComfyUI_repackaged · Hugging Face
LoRA · 304 MB · SAFETENSORS · Hugging Face · Kijai/WanVideo_comfy
Node pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry