Wan

Free plan LinuxSelf-hosted

What Wan is

Wan is an open suite of video foundation models for generating and editing media. It supports text-to-video and image-to-video, as well as video editing, text-to-image, and video-to-audio tasks. The models can place Chinese or English text inside generated videos. For video editing and generation, VACE can take a prompt and optional video, mask, and image inputs. Wan-VAE handles encoding and decoding 1080P video of any length while preserving temporal information.

Wan runs on Linux and can be self-hosted. The T2V-14B model supports 480P and 720P output; T2V-1.3B supports 480P, and its 720P results are described as less stable. The T2V-1.3B model requires 8.19 GB of VRAM. The project reports about four minutes to generate a five-second 480P video on an RTX 4090 without quantization. Model downloads are available through Hugging Face and ModelScope. Installation guidance covers pip and Poetry, with PyTorch 2.4.0 or later required by the quickstart. The repository includes command-line and local Gradio inference examples. Wan is free and open source under Apache-2.0, and commercial use is permitted. Integrations include Diffusers and ComfyUI; prompt extension can use the DashScope API or a local Qwen model.

Who it suits

Wan suits creators and developers who want to run video generation models on Linux or self-host them. It is also relevant to users who need several media-generation tasks, including video editing and text rendered within video, and have the required GPU resources.

What it does well

Supports text-to-video, image-to-video, editing, text-to-image, and video-to-audio.
Can generate Chinese and English text within videos.
T2V-14B supports output up to 720P.
Open source under Apache-2.0; commercial use is permitted.
Integrates with Diffusers and ComfyUI.

What to know first

T2V-1.3B requires 8.19 GB of VRAM.
T2V-1.3B supports 480P; its 720P results are less stable.
The quickstart requires PyTorch 2.4.0 or later.

Limits

The T2V-1.3B model supports 480P, and its 720P results are less stable. The T2V-14B model supports 480P and 720P.

Getting started

Open the Wan2.1 GitHub repository.
Download model weights through Hugging Face or ModelScope.
Use the installation guide's pip or Poetry method, with PyTorch 2.4.0 or later.
Run inference from the command line or start the local Gradio interface.
Use a local Qwen model or the DashScope API for prompt extension.

Verdict

Choose Wan if you want an open source, self-hosted video model suite with multiple generation and editing tasks, and can meet its setup and hardware requirements. Look elsewhere if you need a hosted service or higher resolution from the T2V-1.3B model.

Questions

Is Wan free?

Yes. Wan is free and open source under the Apache-2.0 license.

Can Wan be used commercially?

Yes, commercial use is permitted.

Which platforms does Wan support?

The listed platforms are Linux and self-hosted deployment.

What video tasks does Wan support?

It supports text-to-video, image-to-video, video editing, text-to-image, and video-to-audio.

What resolutions are supported?

T2V-14B supports 480P and 720P; T2V-1.3B supports 480P, with less stable results at 720P.

How can I install Wan?

The guide documents pip and Poetry installation. The quickstart calls for PyTorch 2.4.0 or later and includes command-line and local Gradio examples.

Facts from Wan's own pages

Commercial use: Yes · github.com, 19 Sept 2026
Downloads: Model weights are linked for download through Hugging Face and ModelScope. · github.com, 8 Oct 2026
Generation tasks: The models support text-to-video, image-to-video, video editing, text-to-image, and video-to-audio tasks. · github.com, 8 Oct 2026
GPU requirement: The T2V-1.3B model requires 8.19 GB VRAM according to the repository, which also reports about four minutes to generate a five-second 480P video on an RTX 4090 without quantization. · github.com, 8 Oct 2026
Hardware requirement: The T2V-1.3B model requires 8.19 GB of VRAM, and the project says it can generate a five-second 480P video on an RTX 4090 in about four minutes without quantization. · github.com, 8 Oct 2026
Image and video tasks: The repository lists image-to-video, video editing, text-to-image, and video-to-audio among its supported tasks. · github.com, 8 Oct 2026
Image-to-video: Yes · github.com, 19 Sept 2026
Installation: The installation guide documents pip and Poetry installation methods. · github.com, 8 Oct 2026
Integrations: The project says Wan2.1 T2V and I2V were integrated into Diffusers and that Wan2.1 was integrated into ComfyUI. · github.com, 8 Oct 2026
License: The repository identifies its license as Apache-2.0. · github.com, 8 Oct 2026
Local interface: The repository includes instructions for running a local Gradio interface. · github.com, 8 Oct 2026
Maximum resolution: 720p · github.com, 19 Sept 2026
Maximum video length: 0.083333 · github.com, 19 Sept 2026
Model downloads: The project provides model downloads through Hugging Face and ModelScope. · github.com, 8 Oct 2026
Product: Wan2.1 is an open suite of video foundation models for video generation. · github.com, 8 Oct 2026
Prompt extension: Prompt extension can use the DashScope API or a local Qwen model. · github.com, 8 Oct 2026
Requirements: The quickstart says to ensure PyTorch 2.4.0 or later before installing dependencies. · github.com, 8 Oct 2026
Resolutions: The repository lists T2V-14B at 480P and 720P, and T2V-1.3B at 480P; it cautions that 1.3B results at 720P are less stable than at 480P. · github.com, 8 Oct 2026
Resolution support: The T2V-14B model supports 480P and 720P, while T2V-1.3B supports 480P. · github.com, 8 Oct 2026
Runtime: The quickstart instructs users to install dependencies with PyTorch 2.4.0 or later and provides command-line and local Gradio inference examples. · github.com, 8 Oct 2026
Support: The repository directs users to Discord and WeChat groups to contact the research or product teams. · github.com, 8 Oct 2026
Text generation: Wan2.1 can generate Chinese and English text within videos. · github.com, 8 Oct 2026
Text to video: Wan2.1 supports text-to-video generation with 1.3B and 14B models. · github.com, 8 Oct 2026
Video editing: VACE accepts a text prompt and optional video, mask, and image inputs for video generation or editing. · github.com, 8 Oct 2026

Wan