Du hast einen ComfyUI-Workflow? Erstelle direkt im Browser eine verständliche Workflow-Dokumentation.Workflow verständlich machen.Jetzt ausprobieren ↗
An image to video ComfyUI workflow with the video model Wan Vace
Laufzeitprofil
Quellenbeschreibung
Description
An image to video ComfyUI workflow with the video model Wan Vace
Howto
Drag an initial image into the image node, adjust the prompt, and press queue. The rest should fit.
Description
An image to video ComfyUI workflow with the video model Wan Vace and a special lora called Causvid. Which reduces the creation time dramatically. It is based at the default template that you can find in ComfyUI. But with some modifications.
Note that upscaling better happens in an extra step. You can find example upscaling workflows in my article about video upscaling in comfyui:
KI-generierter Kommentar
KI-generierte Erklärung auf Grundlage von Quellen- und Konfigurationsdetails. Vorschläge sind klar gekennzeichnet.
Use Wan 2.1 Image to Video in ComfyUI to turn an initial image and adjusted prompt into a video with Wan Vace.
Drag an initial image into the image node, adjust the prompt, and press queue to generate a video.
The described examples are 480p, with upscaling handled as a separate extra step.
Plan for at least 12 GB of VRAM and 32 GB of system RAM; the workflow was generated with 16 GB of VRAM.
Listed model files include umt5_xxl_fp16.safetensors, wan_2.1_vae.safetensors, wan2.1_vace_1.3B_fp16.safetensors, and wan2.1_vace_14B_fp16.safetensors.
Listed LoRA files are Wan21_CausVid_14B_T2V_lora_rank32.safetensors and Wan21_CausVid_bidirect2_T2V_1_3B_lora_rank32.safetensors.
Listed node packs include comfyui-frame-interpolation, comfyui-kjnodes, and comfyui-videohelpersuite.
The description includes CausVid LoRA examples with the LoRA off and on.
The description labels these as example times: 67 minutes with the 14B model and CausVid off, 10 minutes with the 14B model and CausVid on, and 3 minutes with the 1.3B model and CausVid off.
Expand the collapsed Note nodes to read their additional information.
Vorschlag · nicht geprüft
Before running, verify that the listed model files and node packs are available in ComfyUI.
Vorschlag · nicht geprüft
The description reports that a setup with 16 GB of VRAM and 32 GB of RAM could not create the 720p version; verify your hardware before attempting 720p.
Vorschlag · nicht geprüft
Muss das bearbeitet werden?
Melden Sie sich an, um eine Änderungsanfrage zu senden.
Quellen
1 QuelleQuellenauszüge
1 AuszugSource context: 2034 downloads · Type Workflows · Base model Wan Video 14B i2v 480p
There are some collapsed Note nodes besides the important nodes. Click at them to expand them. Some contains valuable informations.
Time
The example video in 480p , the 14b model and the causvid lora off took 67 minutes.
The example video in 480p , the 14b model and the causvid lora on took 10 minutes.
The example video in 480p , the 1.3b model and the causvid lora off took 3 minutes.
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. I was not able to create the 720p version of the video with just 16 gb vram and my 32 gb ram.
Geschätzter VRAM Bedarf
Schätzung nicht verfügbar
391 MB über 2 von 6 Modell-Dateien. Gesamtmodell-Dateien + 25% Ladeaufwand + 2 GB Ausführungs-Puffer, aufgerundet.
Anforderungen
9 AnforderungenText 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_vace_1.3B_fp16.safetensors
Nicht aufgelöstCheckpoint · Unknown
wan2.1_vace_14B_fp16.safetensors
Datei ungeprüftCheckpoint · 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
LoRA · 87.0 MB · SAFETENSOR · Unknown
Knotenpaket · Registry
Knotenpaket · Registry
Knotenpaket · Registry
