MooshieUI

MooshieUI is a beginner-friendly interface for ComfyUI that runs in two modes:

• Desktop app via Tauri (Windows/Linux, macOS source build)
• Browser/server mode via the built-in web server (LAN/Docker friendly, mobile UI)

Built with Svelte 5 + Rust, it hides ComfyUI's node-graph complexity behind a clean, guided workflow so you can generate without hand-editing graphs.

MooshieUI is free and open source. If it saves you time or sparks joy, a sponsorship keeps the updates coming. No pressure, just gratitude. 🙏 (where does the money go?)

---

📚 Documentation

Full guides live in the MooshieUI Wiki:

Guide	Covers
Installation	Desktop, Docker, remote/cloud ComfyUI, macOS source build
Generation Basics	Modes, generation controls, dimensions
Prompting Guide	Autocomplete, presets, wildcards, interrogation
Prompt Assistant	LLM-assisted prompt building
Models & the Model Hub	Supported architectures, auto-detection, downloads
Upscaling & Face Fix	Tiled diffusion, guidance nodes, face fix
ControlNet & Style Transfer	ControlNet and reference/style transfer
Inpainting & the Canvas Editor	Mask painting and selective edits
Compare Grid	XYZ parameter sweeps
Gallery & Metadata	Persistent gallery, metadata import/remix
Server, LAN & Multi-User	Self-hosting, roles, auth, mobile
Settings & Accessibility	Persistence, i18n, accessibility
FAQ	Common questions

---

✨ Highlights

• Three generation modes - text to image, image to image, and inpainting with a built-in canvas/mask editor; settings carry over between modes.
• Full generation controls - searchable checkpoint/VAE/LoRA pickers with auto-download, all ComfyUI samplers and schedulers, steps/CFG/seed/batch, and smart dimension presets.
• Smart model detection - 20+ architectures (SD 1.5, SDXL, Illustrious/NoobAI, Pony, SD3/3.5, the Flux family, Chroma, Z-Image, Wan, Qwen, AuraFlow, PixArt, HunyuanDiT, Stable Cascade, Kolors, Anima, Mugen, Nanosaur) identified by SHA256 hash, each with auto-applied sampler/scheduler/CFG presets.
• Tiled-diffusion upscaling - MultiDiffusion/SpotDiffusion with anti-hallucination guidance nodes, one-click upscale, and YOLOv8 face fix.
• Compare Grid (XYZ) - per-cell parameter sweeps stitched into a single labelled image.
• Real-time feedback - live latent previews, progress phases, and cancel, streamed over WebSocket.
• Gallery & metadata - persistent SQLite-backed gallery; drag a PNG back in to restore its settings (SwarmUI/A1111 metadata + stealth alpha).
• Self-hostable - headless web server with roles, per-user galleries, auth, and a dedicated mobile layout.
• 11 languages - 2,000+ translation keys with full locale parity, switchable without restart.

See the Wiki for the full feature reference.

---

📦 Quick Start

Desktop (Windows/Linux)

1. Download a release from Releases.
2. Run the app. The setup wizard downloads uv, Python, ComfyUI, and PyTorch (GPU auto-detected) and installs MooshieUI's custom nodes - no Python or pip setup required.
3. Start generating; ComfyUI launches automatically.

~5–10 GB disk, 5–15 minutes on first launch. macOS, Docker, and remote/cloud ComfyUI setups are covered in Installation.

Self-host (Docker)

cp .env.example .env   # optional: credentials/ports
docker compose up -d --build

Open http://localhost:3200 (or your configured MOOSHIEUI_PORT). Full server/LAN/multi-user setup: Server, LAN & Multi-User.

Build from source

git clone https://github.com/Mooshieblob1/MooshieUI.git
cd MooshieUI
npm install
npm run tauri dev      # hot-reload dev
npm run tauri build    # production build

---

🏗️ How it works

1. You adjust settings in the Svelte UI.
2. On Generate, settings go to the Rust backend via the IPC bridge (ipcInvoke() on desktop, HTTP/SSE in browser mode).
3. Rust builds a ComfyUI workflow JSON from templates - no node graph exposed.
4. The workflow is submitted to ComfyUI's /prompt API.
5. WebSocket streams progress and previews back to the UI in real time.

MooshieUI also ships custom ComfyUI nodes (tiled diffusion, soft/smart guidance, an SDXL↔Flux2 VAE adapter, Nanosaur DiT support, and face fix) that are auto-installed into ComfyUI. Details live in Models & the Model Hub. The tiled diffusion node is also available as a standalone ComfyUI custom node: ComfyUI-MooshieTiledDiffusion.

---

🛠️ Tech Stack

Layer	Technology
Frontend	Svelte 5, TypeScript 6, Tailwind CSS 4
Runtime	Tauri desktop app + axum headless web server
State	Svelte 5 runes - class-based singleton stores
Persistence	Tauri Store (JSON) + SQLite (rusqlite)
ComfyUI transport	REST + WebSocket via Rust (reqwest, tokio-tungstenite)
Inference	ONNX Runtime (ort) for WD EVA02 image interrogation
Autocomplete	Danbooru + Anima tag databases (~140k tags)
i18n	11 languages, 2,000+ keys, runtime switching
Build	Vite 6 + @sveltejs/vite-plugin-svelte

---

🔒 Security

Automated GlassWorm resistance checks run on every push and pull request to catch supply-chain attacks that hide payloads in invisible Unicode variation selectors or tamper with git timestamps. The CI workflow (.github/workflows/glassworm-scan.yml) blocks merges on failure. Contributors should enable the same checks locally:

bash scripts/setup-hooks.sh

---

💛 Support & Where the Money Goes

First off, to be clear: this is not meant to be income. MooshieUI is a passion project. I build it in my spare time around a regular day job, I don't expect to earn anything from it, and right now the running costs come straight out of my own pocket.

If you sponsor the project, here is exactly where it goes:

• Domain & hosting - keeping the project site and download links online.
• SaaS & dev tooling - the paid services and tools used to actually build and ship MooshieUI.
• GitHub Pro+ - CI/CD minutes for the build, release, and security-scan pipelines.

The goal is simply to stop the project from costing me money to keep alive. Anything beyond covering costs just goes right back into building more features, faster.

Longer term, the ideal is that MooshieUI can outlast my own availability. I intend to support this project for as long as I can, but every maintainer has lulls, and life can pull you away for a stretch. A small buffer means the domain, hosting, and infrastructure stay paid up through those quiet periods, so the project stays online and usable even when I am not actively maintaining it.

Sponsoring is completely optional and the app will always be free and open source either way. Thank you for even considering it. 🙏

---

🤝 Contributing

Pull requests are welcome. main is protected: open a PR from a chore/<topic> branch after local validation and GlassWorm pre-commit checks. See push-instructions.md for the full workflow (branch naming, build gates, IPC/gallery conventions, and CI).

---

📋 Changelog

See CHANGELOG.md for the full version history.

---

📄 License

Licensed under the GNU Affero General Public License v3.0.

---

🙏 Acknowledgments

MooshieUI stands on the shoulders of a huge amount of open-source work. Sincere thanks to every project, researcher, model creator, and service below.

Core foundations

• ComfyUI (comfyanonymous) - the diffusion backend that powers all generation. MooshieUI would not exist without it.
• Tauri - the Rust desktop app framework, plus its store, shell, dialog, fs, clipboard, updater, and process plugins.
• Svelte, Tailwind CSS, Vite, and TypeScript - the frontend stack.
• PyTorch - the ML framework behind ComfyUI inference.
• uv (Astral) - manages Python and the ComfyUI environment during setup.

Inference runtimes

• llama.cpp (ggml-org) - local LLM inference for the Prompt Assistant.
• ONNX Runtime (Microsoft) via the ort Rust crate - runs the image interrogator.
• Ultralytics - YOLOv8/YOLO11 detection powering Face Fix and segment refinement.

Bundled third-party ComfyUI nodes

Auto-installed into ComfyUI alongside MooshieUI's own nodes:

• comfyui_controlnet_aux (Fannovel16) - ControlNet preprocessors (Canny, Depth, OpenPose, LineArt, and more).
• ComfyUI-Anima-LLLite (kohya-ss) - LLLite ControlNet support for Anima.
• ComfyUi-Untwisting-RoPE and ComfyUi-Scale-Image-to-Total-Pixels-Advanced (BigStationW) - training-free style transfer (the Anima style-transfer workflow is ported from Untwisting-RoPE's examples) and advanced image scaling.

Research implemented by MooshieUI's own nodes

• MultiDiffusion - Bar-Tal et al., ICML 2023 (arXiv:2302.08113) - overlapping-tile fusion for tiled diffusion upscaling.
• SpotDiffusion - Frolov et al., 2024 (arXiv:2407.15507) - seam-free shifted-window tiling.
• CFG Rescale - "Common Diffusion Noise Schedules and Sample Steps are Flawed" (Lin et al., arXiv:2305.08891) - the basis of MooshieSoftGuidance, plus community "Mahiro"-style positive-biased guidance behind MooshieSmartGuidance.
• OmniSR - "Omni Aggregation Networks for Lightweight Image Super-Resolution" (CVPR 2023, arXiv:2304.10244) - lightweight super-resolution.

Models & model creators

• WD EVA02 Large Tagger v3 (SmilingWolf) - image interrogation/tagging.
• CLIPSeg (CIDAS) - text-prompted region detection for <segment:...> refinement.
• Face detection models - Anzhc's YOLOs (default face segmentation) and ADetailer models (Bingsu) for Face Fix.
• Upscalers - OmniSR, SPAN, and DAT (IllustrationJaNai) model weights hosted by Acly and AshtakaOOf.
• Prompt Assistant LLMs - Qwen (Alibaba) instruct models and DanTagGen (KBlueLeaf), with GGUF quantizations by bartowski.
• Supported architectures & recommended models - Anima (Circlestone Labs), Mugen (CabalResearch), Nanosaur (whose VAE builds on Meta's DINOv3 and whose text encoder uses Google's Gemma 3), SDXL and its VAE (Stability AI), and Juice (Enferlain).

Ecosystem compatibility & inspiration

• SwarmUI - MooshieUI reads and writes SwarmUI-compatible metadata, supports its <segment>/<fromto> prompt syntax, and borrows its backend-handler and in-memory image delivery patterns.
• AUTOMATIC1111 Stable Diffusion WebUI - legacy metadata parsing and the (tag:1.1) weight syntax.
• InvokeAI and NovelAI - additional prompt weight syntaxes MooshieUI understands and converts.
• stealth-pnginfo (ashen-sensored) - the alpha-channel metadata embedding technique.
• ComfyUI Impact Pack (ltdrdata) - the face-detailer concept that MooshieUI's lightweight FaceDetailer node reimplements.

Data & services

• CivitAI - model search, hash lookup, and metadata.
• Hugging Face - hosting for nearly every model MooshieUI downloads.
• Danbooru and Gelbooru - the tag taxonomies behind autocomplete (~140k tags; Gelbooru-derived Anima list curated by BetaDoggo).
• Animadex - the character and LoRA database integration.
• Photopea - the embedded full image editor.
• GitHub and Cloudflare - code hosting, CI/CD, releases, and the CDN behind the artist gallery.

Libraries

• Frontend: Konva + svelte-konva (canvas editor), marked (markdown), DOMPurify (sanitization), SortableJS (drag & drop), and ntc-ts (a TypeScript port of Chirag Mehta's "Name that Color").
• Rust: Tokio, axum, reqwest, tokio-tungstenite, rusqlite + SQLite, jxl-oxide and jxl-encoder (JPEG XL gallery storage), and RustCrypto's Argon2 (password hashing).

If your work is used in MooshieUI and you feel it isn't credited properly here, please open an issue, it will be fixed promptly.