Wan2GP
App : A product you use directly
Interface to run Wan, Hunyuan and LTX on modest GPUs.
The simplest way to test open source video models without ComfyUI.
Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.
What it does, concretely
Generating videos, images or audio with open source models usually requires a complex technical setup, often reserved for experienced users. Wan2GP (now called WanGP) provides a unified web interface that lets you run these models directly from a browser, without going through tools like ComfyUI.
The application installs locally on Windows or Linux and runs entirely on your own machine. It supports a wide range of models for video (Wan 2.1/2.2, MiniMax H3, LTX, Hunyuan Video), image (Flux, Krea 2, HiDream, Ideogram 4) and audio or text-to-speech (Qwen3 TTS, Ace Step, Chatterbox). Models are downloaded automatically based on your hardware, and several quantized formats help reduce video memory requirements.
WanGP is aimed at people who want to experiment with open source generative models without owning a high-end GPU: some models run with as little as 6 GB of VRAM, and Nvidia cards from the GTX 10XX series onwards as well as some AMD cards are supported. A generation queue, a headless mode and an API also enable more advanced workflows.
However, the tool is not suited to those looking for a ready-to-use online service: you need to install the application, manage updates and have a sufficiently powerful computer. macOS users are not supported.
What it offers
- ·Video, image and audio generation from a local web interface, with no account or subscription required.
- ·Support for modest GPUs with as little as 6 GB of VRAM, including Nvidia GTX 10XX, RTX 20XX and some AMD RDNA 2 to 4 cards.
- ·Automatic download of model files suited to your hardware, with support for quantized formats including int8, fp8, gguf and NV FP4.
- ·Generation queue allowing you to line up multiple video, image or audio jobs and retrieve results later.
- ·Built-in preparation and post-processing tools: mask editor, background remover, pose and depth extractor, temporal upsampling (RIFE, DLSS 5) and spatial upsampling.
- ·Deepy offline agent for orchestrating generation, transcription or video-splitting tasks, with a mobile-friendly web app.
- ·LoRA customisation and support for adding your own checkpoints or finetunes from Hugging Face or CivitAI.
- ·WanGP API and headless command-line mode for integrating generative capabilities into other applications or running batch jobs.
In practice
- Price
- Free
- Commercial use
- Commercial use: yes
- Available on
- windows, linux
- Licence
- Apache-2.0
- Runs locally
- Yes, 6 GB VRAM recommended
- FR interface
- No
Security and data
- Rating
- Security: nothing to flag
- Your data
- Data stays local
- Maintenance
- Active
Links and repository checked on 25 October 2026 · added on 1 March 2025