jn46ddlZddlmZGddeZdS)N) WAN2GPPluginc*eZdZfdZdZdZxZS) AboutPlugincrtd|_d|_d|_dS)Nz About Tabz1.0.0z5Credits for the creator and all co-creators of WAN2GP)super__init__nameversion description)self __class__s tjd?tjd@tjdAtjdBtjdCtjdDtjdEtjdFtjdGtjdHtjdIdS)JNz

WanGP - AI Generative Models for the GPU Poor by DeepBeepMeep (GitHub)

zMany thanks to:z- Bria AI for Fibo 2.0 Dev\* image generation and editing, and Fibo 2.0 Edit Turbo\* four-step editing (https://huggingface.co/briaai/Fibo-2.0-dev, https://huggingface.co/briaai/Fibo-2.0-edit-turbo)z- Alibaba Wan Team for the best open source video generators (https://github.com/Wan-Video/Wan2.1, https://github.com/Wan-Video/Wan2.2)z- Alibaba Vace, Multitalk and Fun Teams for their incredible control net models (https://github.com/ali-vilab/VACE), (https://github.com/MeiGen-AI/MultiTalk) and (https://huggingface.co/alibaba-pai/Wan2.2-Fun-A14B-InP) z- Tencent for the impressive Hunyuan Video\* models (https://github.com/Tencent-Hunyuan/HunyuanVideo, https://github.com/Tencent-Hunyuan/HunyuanVideo-1.5)z- Black Forest Labs for the innovative Flux\* image generators, including Flux 3 Image Klein\* and Klein Base\* (https://github.com/black-forest-labs/flux, https://github.com/black-forest-labs/flux-3-image-klein)zt- Alibaba Qwen Team for their state of the art Qwen Image\* generators (https://github.com/QwenLM/Qwen-Image)z- inclusionAI for Ming Image 0.1 Design and Design-Layer (https://github.com/inclusionAI/Ming-Image, https://huggingface.co/inclusionAI/Ming-Image-0.1-Design, https://huggingface.co/inclusionAI/Ming-Image-0.1-Design-Layer)z- OpenMOSS Team for MOSS-Image Base and SFT, with text rendering and layout-guided image generation (https://github.com/OpenMOSS/MOSS-Image, https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base, https://huggingface.co/OpenMOSS-Team/MOSS-Image-SFT)z- Krea AI for the Krea 2\* RAW and Turbo image generators (https://github.com/krea-ai/krea-2, https://huggingface.co/krea/Krea-2-Raw, https://huggingface.co/krea/Krea-2-Turbo, https://www.krea.ai/blog/krea-2-technical-report)z- Lightricks for their super fast LTX Video\* models and LTX-2\* audio-video model (https://github.com/Lightricks/LTX-Video, https://github.com/Lightricks/LTX-2)z{- MiniMax AI for the MiniMax H3\* native audio-video generation models (https://huggingface.co/MiniMaxAI/MiniMax-H3)a- NVIDIA Research Sol-Attn team for Sol-Attn and its reference kernels, and Saganaki22 (drbaph) for the consumer-GPU Triton kernel improvements and MiniMax H3 integration (https://nvlabs.github.io/Sana/Sol-Attn/, https://github.com/Saganaki22/ComfyUI-sol-attn/tree/e2fc225)z- Echo Team @ Joy Future Academy, JD for JoyAI-Echo\* (https://github.com/jd-opensource/JoyAI-Echo, https://huggingface.co/jdopensource/JoyAI-Echo)zm- Eyeline Labs, Netflix and collaborators for Vista4D (https://github.com/Eyeline-Labs/Vista4D)z- Alibaba PAI for Wan2.1-Fun Control-Camera (https://github.com/aigc-apps/VideoX-Fun) and Alibaba DAMO Academy for Uni3C camera control (https://github.com/alibaba-damo-academy/Uni3C)z- Tencent Hunyuan3D Team for the Hunyuan3D 2.1\* image-to-3D shape and texture models (https://github.com/Tencent-Hunyuan/Hunyuan3D-2.1); license and notice in models/hunyuan3d/LICENSE and Notice.txtz- Tencent HY-World Team for WorldMirror 2.0\* 3D reconstruction from photos or videos (https://github.com/Tencent-Hunyuan/HY-World-2.0); license and notice in models/worldmirror/LICENSE and Notice.txta)- Tencent Hunyuan for the Hy-Nomas explorable world model (https://github.com/Tencent-Hunyuan/Hy-Nomas), with NVIDIA's PixelUMM\* video decoder (https://github.com/nv-tlabs/PixelUMM, non-commercial weights); licenses and notice in models/hy_nomas/LICENSE, LICENSE_PixelUMM and NOTICEa%- Alaya Lab / The Evoke Team for EVOKE-Turbo explorable worlds with a geometric memory (https://github.com/AlayaLab/Evoke), with aigc3d's ViGeo1.1\* video depth model (https://github.com/aigc3d/ViGeo, non-commercial weights); license and notice in models/evoke/LICENSE and NOTICEz- Microsoft for MoGe-2 (https://github.com/microsoft/MoGe), Meta for MapAnything with CMU AirLab's UniCeption (https://github.com/facebookresearch/map-anything): 3D reconstruction from photos and videosz- Microsoft for TRELLIS.2 and the TRELLIS sparse-structure decoder (https://github.com/microsoft/TRELLIS.2), with Meta's DINOv3\* image encoderz- VAST / Tripo for TripoSplat image-to-Gaussian-splats (https://github.com/VAST-AI-Research/TripoSplat), with Meta's DINOv3\* image encodera8- VAST-AI Research and Yuchen Lin for TripoSG\*, PartCrafter\* and GeoSAM2 image-to-mesh, part generation and 3D part segmentation (https://github.com/VAST-AI-Research/TripoSG, https://github.com/wgsxm/PartCrafter); TripoSG code includes Tencent Hunyuan-derived parts, see models/triposg/Notice.txtz- Zijie Wu, Chaohui Yu, Fan Wang and Xiang Bai (HUST / Alibaba DAMO Academy) for AnimateAnyMesh text-driven mesh animation (https://github.com/JarrentWu1031/AnimateAnyMesh)z^- Resemble.AI for the incredible ChatterBox (https://github.com/resemble-ai/chatterbox)zd- HeartMuLa Team for the open music generation models (https://github.com/HeartMuLa/heartlib)a%- Multimodal Art Projection (MAP) Team and collaborators for YuE2\* song generation, SheetSage2\* music transcription and the MERT2\* music encoder (https://github.com/multimodal-art-projection/YuE, https://huggingface.co/m-a-p/SheetSage2, https://huggingface.co/m-a-p/MERT-v2-FullSong)a- **Mothersuperior** for the [YuE2 Hum-to-Song adapter](https://huggingface.co/Mothersuperior/YuE2-hum-to-song)\*, hum-conditioning workflow and [real-audio acoustic LoRA](https://huggingface.co/Mothersuperior/yue2-mothersuperior-realaudio-tokenizer-v4)\* used by this contribution.z- ACE-Step Team for the ACE-Step music generation model (https://github.com/ace-step/ACE-Step) & ACE-Step 1.5 (https://github.com/ace-step/ACE-Step-1.5)z[- Stability AI for Stable Audio 3\* (https://github.com/Stability-AI/stable-audio-3)zO- Alibaba Qwen Team for Qwen 3 TTS (https://github.com/QwenLM/Qwen3-TTS)z- Fish Audio for Drama 3\* expressive speech, dialogue and voice cloning, including the INT4 Experts variant (https://github.com/fishaudio/fish-speech, https://huggingface.co/fishaudio/drama-3)zv- Tencent Hunyuan Team for the AuK audio generation and editing models (https://github.com/Tencent-Hunyuan/AuK)zY- KugelAudio for Kugel Audio (https://huggingface.co/kugelaudio/kugelaudio-0-open)zZ- Remade_AI : for their awesome Loras collection (https://huggingface.co/Remade-AI)zR- ByteDance Bernini Team for Bernini (https://github.com/bytedance/Bernini)z- ByteDance : for their great Wan and Flux extensions Lynx (https://github.com/bytedance/lynx), UMO (https://github.com/bytedance/UMO), USO (https://github.com/bytedance/USO) z- Hugging Face for providing hosting for the models and developing must have open source libraries such as Tranformers, Diffusers, Accelerate and Gradio (https://huggingface.co/)zK
Huge acknowledgments to these great open source projects used in WanGP:zJ- Rife: temporal upsampler (https://github.com/hzwer/ECCV2022-RIFE)zN- DwPose: Open Pose extractor (https://github.com/IDEA-Research/DWPose)z- DepthAnything\* & Midas: Depth extractors (https://github.com/DepthAnything/Depth-Anything-V2) and (https://github.com/isl-org/MiDaSzL- Matanyone\*: Mask Generation (https://github.com/pq-yang/MatAnyone)z- SAM2 and SAM3\*: Mask Generation (https://github.com/facebookresearch/sam2, https://github.com/facebookresearch/sam3)zS- Pyannote: speaker diarization (https://github.com/pyannote/pyannote-audio)zY- Sherpa-ONNX: speaker diarization runtime (https://github.com/k2-fsa/sherpa-onnx)zC- Kokoro: text-to-speech (https://github.com/hexgrad/kokoro)z- OpenPhonemizer (NeuralVox) and DeepPhonemizer (Axel Springer Ideas Engineering): English phonemizer for Kokoro's out-of-dictionary words (https://github.com/NeuralVox/OpenPhonemizer, https://github.com/as-ideas/DeepPhonemizer)z- SeedVC\*: voice conversion (https://github.com/Plachtaa/seed-vc), downloaded on first use from its separate GPL-3.0 repository (https://github.com/deepbeepmeep/WanGP-SeedVC)zK- MMAudio\*: sound generator (https://github.com/hkchengrex/MMAudio)z- model-viewer, three.js and Spark: 3D asset and Gaussian splat viewers (https://github.com/google/model-viewer, https://github.com/mrdoob/three.js, https://github.com/sparkjsdev/spark)z- resvg (Yevhenii Reizner, linebender) and resvg-py: SVG rendering for Deepy's drawings (https://github.com/linebender/resvg, https://github.com/baseplate-admin/resvg-py)z- ddgs (deedy5) and primp: metasearch over public engines and HTTP client for Deepy's web search (https://github.com/deedy5/ddgs, https://github.com/deedy5/primp)ze- pypdf: PDF text for Deepy's web reading and Python sandbox (https://github.com/py-pdf/pypdf)z- Playwright: web page screenshots for Deepy, through the installed Edge or Chrome (https://github.com/microsoft/playwright-python)z- ByteDance Seed Team and the SeedVR2 authors: one-step image and video restoration (https://github.com/ByteDance-Seed/SeedVR, https://arxiv.org/abs/2506.05301)zc- FlashVSR: high quality video super-resolution (https://github.com/OpenImagingLab/FlashVSR)z_- NVIDIA PiD\*: diffusion-based image super-resolution (https://github.com/nv-tlabs/PiD)z- Carasibana: the MIT-licensed ComfyUI H3 FaceRefine workflow adapted by WanGP (https://github.com/Carasibana/ComfyUI-H3-FaceRefine)z- Ultralytics YOLO\* and Bingsu's ADetailer: face and person detection used by H3 Face Refiner (https://github.com/ultralytics/ultralytics, https://github.com/Bing-su/adetailer)z- InsightFace\*: face detection, landmarks, and identity embeddings used by H3 Face Refiner (https://github.com/deepinsight/insightface)z- Merserk and DLSS5-Feeder: foundations for the DLSS 5 video-processing integration (https://github.com/Merserk/dlss5-visual-enhancer, https://github.com/jlrouzies-fr/DLSS5-Feeder)zM
Special thanks to the following people for their Contributions & Support:z- Tophness : Designed & developped the Queuing Framework, Edit Mode, Windows Installation scripts and WanGP PlugIns Systemzo- GKArtist : For the great Wan2GP Desktop installer (https://github.com/GKartist75/Wan2GP-Desktop-Tauri)z]- Gunther-Schulz : for adding image Start Image / Image Refs storing in Video metadatazK- Cocktail Peanuts : QA and simple installation via Pinokio.computerz@- huangyebiaoke : for his support in porting WanGP to MPSzV- AmericanPresidentJimmyCarter : added original support for Skip Layer GuidancezI- Reevoy24 : for his repackaging / completion of the documentationz
\* Restrictive license (non-commercial, territorial, usage-based or copyleft terms): refer to and comply with the corresponding license terms.)grMarkdownrs rrzAboutPlugin.create_about_uis' b c c c %&&& f g g g e f f f z { { { y z z z u v v v L M M M | } } } U V V V @ A A A A B B B S T T T v w w w r s s s D E E E \ ] ] ] f g g g g h h h A B B B } ~ ~ ~ v w w w n o o o j k k k Q R R R J K K K tuuu z{{{  @ @ @ s t t t v w w w rsss efff ` a a a M N N N oppp pqqq hiii M N N N P Q Q Q abbb `aaa deee l m m m cddd ] ^ ^ ^ ijjj oppp YZZZ I J J J N O O O bccc e f f f O P P P G H H H {||| a b b b ~    yzzz vwww b c c c P Q Q Q g h h h Y Z Z Z cddd X Y Y Y F G G G sttt abbb VWWW lmmm _``` j k k k k kr)__name__ __module__ __qualname__rrr __classcell__)r s@rrrsiSSSSS    LkLkLkLkLkLkLkrr)gradiorshared.utils.pluginsrrrrr$sk------ZkZkZkZkZk,ZkZkZkZkZkr