Skip to content
OpenMicroDrama

deepbeepmeep /Wan2GP

Desktop app that runs open video, image and audio models on your own PC
Model toolingCustom licenseEnglish README
Open on GitHubSave
Stars
9.8k
30 days
+32
Last push
yesterday

About Wan2GP

WanGP runs open video, image and audio models on your own computer. Built for older or low-VRAM GPUs, it picks model files your card can handle. You get a browser UI, galleries and a job queue. Its license isn't standard, so check the terms before you depend on it.

Best for: People with older or low-VRAM GPUs who want open video, image and audio models on their own machine.

What it does

  • Runs open video models like Wan 2.1/2.2, MiniMax H3, LTX-2.x and Hunyuan Video
  • Runs some models on as little as 6 GB of VRAM
  • Works on older Nvidia cards (GTX 10XX, RTX 20XX) and AMD RDNA 2-4 GPUs
  • Fetches the checkpoint files that match your GPU automatically
  • Queues video, image and audio jobs, or runs batches from the command line
  • Adds soundtracks with MMAudio and swaps voices with SeedVC

Quickstart

Rewritten from the README. Check the repo for the latest steps.

  1. 1
    Read the setup steps in docs/INSTALLATION.md for your OS and GPU.
  2. 2
    Start the app and open the web interface in your browser.
  3. 3
    Pick a model; WanGP downloads the checkpoint files your hardware needs.
  4. 4
    Queue video, image or audio jobs in the UI, or run batches in headless mode.

Models and languages

Models it works with

  • Wan 2.1/2.2
  • MiniMax H3
  • LTX-2/2.3/2.5
  • Hunyuan Video 1/1.5
  • LongCat
  • Kandinsky
  • Qwen Image
  • Flux 1/2
  • Krea 2
  • Index TTS2/2.5

Interface and docs

  • English

Head-to-head

Alternatives

Other model tooling projects people compare with Wan2GP.

ModelTC

LightX2V

Loads open image and video models onto your own GPU

Model tooling
  • MiniMax H3
  • Wan 2.2
  • Wan 2.1
  • +7
2.9k GitHub stars-2/30dyesterdayApache-2.0

modelscope

DiffSynth-Studio

Trains and runs image, video and audio diffusion models on your own GPU

Model tooling
  • MiniMax H3
  • LTX-2.5
  • Wan-Animate-2
  • +7
13k GitHub stars-1/30d2 days agoApache-2.0

james-see

ltx-video-mac

Mac app that renders AI video with sound using LTX and MiniMax

Model tooling
  • LTX-2
  • LTX-2.3
  • LTX-2.5
  • +4
424 GitHub stars+1/30d24 days agoMIT

wide-trace

open-higgsfield

Self-hosted studio for 38 image and video models

Model tooling
  • Seedance 2.5
  • Seedance 2.0
  • Kling 3
  • +7
3.9k GitHub stars+23/30d15 days agoNo license

Featured on OpenMicroDrama

Do you maintain Wan2GP? Add this badge to your README so English speakers can find our write-up.

Featured on OpenMicroDrama
markdown
[![Featured on OpenMicroDrama](https://openmicrodrama.com/badges/featured.svg)](https://openmicrodrama.com/projects/deepbeepmeep-wan2gp)

The badge links to this page. Want your description changed? Email support@openmicrodrama.com.

Reviewed Oct 1, 2026. Custom WanGP Community License 2.0. Free to run on your own computer; read it before redistributing. Not affiliated with any third-party service using the WanGP name; only the official GitHub repo and wangp.ai / wan2gp.ai are official. We write these descriptions ourselves. The repo's own docs are the final word.