Video models · Alibaba

Wan 2.1 I2V 14B 720P

Name: Wan 2.1 I2V 14B 720P
Author: Alibaba

Backbone: 14.0B
Native output: 1280×720, 81 frames
Default steps: 50
License: apache-2.0
Released: 2025-02

How to run it

Quantized files · GGUF for ComfyUI

Smaller GGUF versions for ComfyUI (with the ComfyUI-GGUF node). No sign-in needed; the model’s license still applies.

huggingface.co/city96/Wan2.1-I2V-14B-720P-gguf (opens in a new tab)

Original weights · Hugging Face

Full-precision files from Alibaba, for fine-tuning, vLLM, Transformers, Diffusers or ComfyUI.

huggingface.co/Wan-AI/Wan2.1-I2V-14B-720P (opens in a new tab)

No sign-in needed.

Official repo · GitHub

Reference code, inference scripts and docs from Alibaba. The weights themselves download from Hugging Face.

github.com/Wan-Video/Wan2.1 (opens in a new tab)

git clone https://github.com/Wan-Video/Wan2.1

Links checked 2 days ago against Hugging Face, Ollama and GitHub.

Apps that run Wan 2.1 I2V 14B 720P

Our pick for each kind of computer, assuming the model fits in its memory. Pick your own GPU or device in the Model Finder for an exact answer.

Mac (Apple silicon)

Draw Things

Native Mac app with its own fast Metal engine and quantized downloads.

Also good:

Windows PC, NVIDIA GPU

ComfyUI

Runs this model with built-in or popular custom nodes; GGUF files load with ComfyUI-GGUF.

Also good:

Linux, NVIDIA or AMD GPU

ComfyUI

Runs this model with built-in or popular custom nodes; GGUF files load with ComfyUI-GGUF.

Also good:

App	Type	Opens it as	Systems	Engine	Start
ComfyUI	Desktop app	Original or GGUF	macOSWindowsLinux	PyTorch, GGUF via ComfyUI-GGUF	Get ComfyUI
Draw Things	Desktop app	Built-in support	macOSiOS	Draw Things	Open in Draw Things
SwarmUI	Desktop app	Original or GGUF	macOSWindowsLinux	ComfyUI	Get SwarmUI
SD.Next	Desktop app	Original or GGUF	macOSWindowsLinux	PyTorch	Get SD.Next
Stability Matrix	Desktop app	Original or GGUF	macOSWindowsLinux	installs ComfyUI, Forge, InvokeAI, SwarmUI, SD.Next or Fooocus	Get Stability Matrix
WanGP	Desktop app	Built-in support	WindowsLinux	PyTorch	Get WanGP

Which apps can open a model depends on the files published for it: GGUF for llama.cpp-based apps, MLX for Apple’s engine, the original weights for vLLM and SGLang. Compare every app.

Memory needed

Peak GPU memory at the native output size. Offloading moves the text encoder, or the whole pipeline, to system RAM between steps.

Precision	Everything on GPU	Text encoder offloaded	Sequential offload	Download
`FP16`	59.3 GB	46.6 GB	28.2 GB	Original weights (opens in a new tab)
`FP8`	38.9 GB	32.6 GB	23.3 GB	Original weights (opens in a new tab)
`Q8 GGUF`	39.9 GB	33.6 GB	23.6 GB	GGUF (opens in a new tab)
`Q4 GGUF`	33.3 GB	27.0 GB	21.3 GB	GGUF (opens in a new tab)

Components

Backbone: 14B
VAE: 0.127B
umT5-XXL: 5.7B
CLIP ViT-H (image): 0.63B

Best hardware for Wan 2.1 I2V 14B 720P

The least expensive GPU in our catalog that runs it without offloading is the Intel Arc Pro B70 (launch price $949) at Q4 GGUF.

Quality	Hardware	Quant	Memory	Speed	Runs
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell	`FP16`	59.3 GB / 96.0 GB	24.4 min	Slow
72Great quality (A) · 72 of 100: how capable the model is at Q4 GGUF, whatever it runs on. Match score 72/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 5090	`Q4 GGUF`	27.0 GB / 32.0 GB	35.1 min	Slowencoder in RAM
76Great quality (A) · 76 of 100: how capable the model is at FP8, whatever it runs on. Match score 63/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GeForce RTX 4090	`FP8`	23.3 GB / 24.0 GB	39.0 min	Slowtight fit · CPU offload
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H100 80GB SXM	`FP16`	59.3 GB / 80.0 GB	23.2 min	Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 80/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H800 80GB SXM	`FP16`	59.3 GB / 80.0 GB	23.2 min	Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H100 80GB PCIe	`FP16`	59.3 GB / 80.0 GB	28.4 min	Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA H100 NVL	`FP16`	59.3 GB / 94.0 GB	25.2 min	Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA GH200 Grace Hopper (96 GB GPU)	`FP16`	59.3 GB / 96.0 GB	23.2 min	Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA RTX PRO 6000 Blackwell Server Edition	`FP16`	59.3 GB / 96.0 GB	25.2 min	Slow
78Great quality (A) · 78 of 100: how capable the model is at FP16, whatever it runs on. Match score 79/100 (quality, fit, speed and memory use) sets the ranking.	NVIDIA A100 80GB PCIe	`FP16`	59.3 GB / 80.0 GB	34.8 min	Slow

Check it on your own hardware

Estimates come from published hardware specs and our open formulas, not from running the model on your machine. Real speed varies with runtime, drivers and settings. Numbers marked “Measured” come from a linked benchmark.