# The ComfyUI alternative that skips the node graph. You wanted a picture. You got a pipeline. ComfyUI on Mac means a bundled Python install that wants at least 4.85 GB of disk, and a node graph between you and the first image. Private Diffusion is one app for iPhone, iPad, and Mac, with its models already tuned for Apple Silicon. No Terminal. No Python. You type what you want, and you get it. [![Pre-order Private Diffusion on the App Store](/app-store/pre-order-badge/en/pre-order.svg)![Pre-order Private Diffusion on the App Store](/app-store/pre-order-badge/en/pre-order-dark.svg)](https://testflight.apple.com/join/DQwfQsEg) ## The Short Version Private Diffusion ComfyUI Picture quality Private DiffusionRetrained after shrinking, one tested build per model ComfyUIWhatever checkpoint and compression you downloaded Speed on Apple Silicon Private DiffusionMLX on the GPU, Core ML on the Neural Engine ComfyUIPyTorch MPS on the GPU alone What leaves your device Private DiffusionCollects no data, not even crash reports ComfyUIPostHog and Datadog telemetry, checked by default The app itself Private DiffusionSandboxed native Swift on iPhone, iPad, and Mac ComfyUIUnsandboxed Electron desktop app ## Better Pictures From Smaller Models Every model is retrained after it's shrunk, so the small version on your device answers the way the full-size original would. How it works Running a big image model on a laptop means shrinking it. In ComfyUI, the quality you get depends on whichever checkpoint and compression you happened to download. Every one of our models is [quantization-aware distilled](https://developer.nvidia.com/blog/how-quantization-aware-training-enables-low-precision-accuracy-recovery/) (QAD): we shrink the weights to 4 bits (6 on Anima and Mage Flow Turbo), then keep training the small model against the full-size original until it answers the same way. You get one tested build per model instead of a guessing game. Sources[NVIDIA: how quantization-aware training and distillation work](https://developer.nvidia.com/blog/how-quantization-aware-training-enables-low-precision-accuracy-recovery/) How We Shrink a Model 1. 1 Full-Size Original The model as its makers trained it, too big for a phone. 2. 2 Shrunk to Fit The weights are cut to a few bits each so the whole model fits on your device. 3. 3 Retrained Against the Original The small model keeps training until it answers the way the full-size one does. Private Diffusion Private Diffusion ships the retrained model. ComfyUI ComfyUI loads whichever checkpoint you found, at whatever compression it came in. ## Faster on Apple Silicon Private Diffusion uses Apple's own frameworks to put the GPU and the Neural Engine to work together. How it works ComfyUI runs on PyTorch's MPS backend, the slowest inference framework on Apple platforms. Its own documentation says [it doesn't use MLX](https://github.com/Comfy-Org/docs/blob/2c665405b557f2d8a36d75dfc41018ec31b18e56/installation/system_requirements.mdx#L69-L72), and with no Core ML path either, the Neural Engine sits idle. Private Diffusion runs the diffusion model on the GPU through MLX and uses the [Neural Accelerators built into the GPU on M5 and M6](https://www.apple.com/newsroom/2026/08/apple-introduces-m6-and-m5-ultra-for-a-big-leap-in-performance-and-ai-compute/), A19 Pro, and A20 Pro. Core ML hands the live preview, the final decode, and the upscaler to the Neural Engine, except for a few of the largest decodes, which come out cleaner on the CPU. Each processor does the part it's best at. Sources[Comfy docs: ComfyUI uses PyTorch MPS, not MLX](https://github.com/Comfy-Org/docs/blob/2c665405b557f2d8a36d75dfc41018ec31b18e56/installation/system_requirements.mdx#L69-L72)[Apple Newsroom: M6 and M5 Ultra, a Neural Accelerator in each GPU core](https://www.apple.com/newsroom/2026/08/apple-introduces-m6-and-m5-ultra-for-a-big-leap-in-performance-and-ai-compute/) Which Processors Do the Work Private Diffusion - GPUUsed - Neural EngineUsed ComfyUI - GPUUsed - Neural EngineNot used ## No Telemetry to Switch Off Private Diffusion collects no data, while ComfyUI's Mac app ships two analytics SDKs with telemetry checked by default. How it works ComfyUI's Mac app, Comfy Desktop, ships [two tracking SDKs, PostHog and Datadog RUM](https://github.com/Comfy-Org/Comfy-Desktop/blob/9c43562d9e1e3a23af13f2e18653935e67bd9eda/src/main/lib/telemetry.ts#L69-L80). The box that turns telemetry on is already checked the first time you launch it, and Comfy's own privacy policy names both vendors. Private Diffusion has nothing to uncheck. It ships without an analytics SDK or a crash reporter, never asks for an account, and collects no data at all. Your prompts and images never reach a Numen server. If you're signed in to iCloud, your gallery mirrors to your own private iCloud. Sources[Comfy Desktop 1.1.6 telemetry source (PostHog, Datadog RUM)](https://github.com/Comfy-Org/Comfy-Desktop/blob/9c43562d9e1e3a23af13f2e18653935e67bd9eda/src/main/lib/telemetry.ts#L69-L80)[Comfy Desktop 1.1.6 first-launch screen: telemetry checked by default](https://github.com/Comfy-Org/Comfy-Desktop/blob/9c43562d9e1e3a23af13f2e18653935e67bd9eda/src/renderer/src/views/FirstUseTakeover.vue#L101)[Comfy Desktop in-app privacy policy naming PostHog and Datadog](https://github.com/Comfy-Org/Comfy-Desktop/blob/9c43562d9e1e3a23af13f2e18653935e67bd9eda/src/renderer/src/lib/legalDocs.ts#L400-L401) ![Comfy Desktop's first-launch screen: a Cloud or Local choice, an unchecked EULA box, and a pre-checked box reading Help improve Comfy by sharing anonymous usage data.](/_astro/comfy-desktop-telemetry-consent.7nxPpshz_1KSYm5.png) Comfy Desktop 1.1.6, first launch. The usage-data box comes checked. Checked October 13, 2026. Comfy Desktop 1.1.6 first-launch screen: telemetry checked by default ## A Native App, Not a Browser in a Box Private Diffusion is a sandboxed, native Swift app built to Apple's Human Interface Guidelines, while Comfy Desktop is an unsandboxed Electron app. How it works Comfy Desktop wraps a web interface in Electron and runs outside the App Sandbox, the wall macOS puts between an app and the rest of your files. Private Diffusion is written in Swift, runs inside the App Sandbox with the hardened runtime on, and follows Apple's Human Interface Guidelines to the T, with Liquid Glass, menu bar commands, and keyboard shortcuts. It also runs on your iPhone and iPad, which ComfyUI does not. SourcesComfy Desktop 1.1.6 code signature: no App Sandbox entitlement ![Comfy Desktop's editor: a node graph of Load Diffusion Model, Load CLIP, two CLIP Text Encode nodes, KSampler, and VAE Decode, with two model loaders outlined in red for missing files.](/_astro/comfy-desktop-mac.DpX-YV8z_ZNqxNa.png) Comfy Desktop 1.1.6 on a Mac, the default workflow before any picture. Checked October 13, 2026. Comfy Desktop 1.1.6 code signature: no App Sandbox entitlement ![Private Diffusion's Studio window on a Mac: a sidebar with Studio, Gallery, Model Library, and API Server, a generated lighthouse image, and a single prompt field with a Generate button.](/_astro/private-diffusion-mac-light.xeGQ_AT8_1Khmbc.png) Private Diffusion Studio on macOS 26, native Swift. ![Private Diffusion's Studio window on a Mac: a sidebar with Studio, Gallery, Model Library, and API Server, a generated lighthouse image, and a single prompt field with a Generate button.](/_astro/private-diffusion-mac-dark.Cub9GTyU_Z1KoSpA.png) Private Diffusion Studio on macOS 26, native Swift. ## Which One Is for You ### Pick ComfyUI If - You want to wire your own node graphs - You run custom nodes - You generate on an NVIDIA card under Windows or Linux ### Pick Private Diffusion If - You want the picture on your Mac or iPhone without setting up Python - You want nothing to uncheck, because nothing is collected - You want a sandboxed native app that also runs on iPhone and iPad ## Private Diffusion and ComfyUI Side by Side How the two apps compare on Apple Silicon, as of the date in the sources below. | What you get | Private Diffusion | ComfyUI | | --- | --- | --- | | How models are shrunk | Retrained after shrinking (QAD) | Depends on the checkpoint you load | | Processors it uses | GPU and Neural Engine | GPU through PyTorch MPS | | Data the app collects | None | Usage and error telemetry, on by default | | Analytics SDKs | None | PostHog and Datadog RUM | | How it's built | Native Swift with Liquid Glass | Electron with bundled Python | | App Sandbox | Yes, with hardened runtime | No | | Runs on | iPhone, iPad, and Mac | Mac, Windows, and Linux; no iPhone or iPad | | Getting started | Download a model, type a prompt | Python install, then a node graph | | Models | Curated, each tuned for your device | Any checkpoint you find | Compare all three How the three apps compare on Apple Silicon, as of the date in the sources below. | What you get | Private Diffusion | Draw Things | ComfyUI | | --- | --- | --- | --- | | How models are shrunk | Retrained after shrinking (QAD) | Compressed after training, no retraining | Depends on the checkpoint you load | | Processors it uses | GPU and Neural Engine | GPU; Neural Engine for select models | GPU through PyTorch MPS | | Data the app collects | None | Purchases, User Content, Identifiers | Usage and error telemetry, on by default | | Analytics SDKs | None | None named in its policy | PostHog and Datadog RUM | | How it's built | Native Swift with Liquid Glass | Mac Catalyst build of its iPad app | Electron with bundled Python | | App Sandbox | Yes, with hardened runtime | Yes | No | | Runs on | iPhone, iPad, and Mac | iPhone, iPad, and Mac | Mac, Windows, and Linux; no iPhone or iPad | | Getting started | Download a model, type a prompt | Pick a model, then tune the settings | Python install, then a node graph | | Models | Curated, each tuned for your device | 248 catalog entries | Any checkpoint you find | ## ComfyUI Alternative FAQ - ### What is the best ComfyUI alternative for Mac? If you want images rather than workflows, Private Diffusion. It installs from the App Store, needs no Python, and runs a curated set of models on Apple Silicon, using MLX on the GPU and Core ML on the Neural Engine. It collects no data. ComfyUI is still the pick for custom node graphs and NVIDIA hardware. - ### Does ComfyUI run well on Mac? ComfyUI runs on Apple Silicon Macs through PyTorch's MPS backend. It does not use MLX, Core ML, or the Neural Engine, and Comfy's own troubleshooting guide suggests falling back to the CPU when MPS causes problems. Private Diffusion is built on Apple's frameworks instead, so the GPU and the Neural Engine work together on every Mac it supports. - ### Can I run ComfyUI on iPhone or iPad? No. ComfyUI's desktop app is available for Mac and Windows, and Comfy Org has no iPhone or iPad app. Private Diffusion runs on iPhone, iPad, and Mac with iOS, iPadOS, or macOS 26.4 or later on Apple Silicon. Bigger models need more memory: Krea 2 Turbo needs an iPad or Mac with 16 GB, while Anima and Flux.2 Klein 4B run on any supported iPhone. - ### Does ComfyUI collect data? Comfy Desktop, ComfyUI's Mac app, ships PostHog for product analytics and Datadog RUM for errors, and its telemetry checkbox is checked by default on first launch. You can uncheck it. Private Diffusion has no analytics SDK and collects no data, so there is nothing to uncheck. ## Make the Picture, Skip the Pipeline Every model trained to hold its quality, on the iPhone, iPad, or Mac you already own. There's no account to make, and nothing you create is sent to us. [![Pre-order Private Diffusion on the App Store](/app-store/pre-order-badge/en/pre-order.svg)![Pre-order Private Diffusion on the App Store](/app-store/pre-order-badge/en/pre-order-dark.svg)](https://testflight.apple.com/join/DQwfQsEg) [See how Private Diffusion compares with Draw Things](/en/draw-things-alternative) ## Sources, Checked October 13, 2026 1. [NVIDIA: how quantization-aware training and distillation work](https://developer.nvidia.com/blog/how-quantization-aware-training-enables-low-precision-accuracy-recovery/) 2. [Apple Newsroom: M6 and M5 Ultra, a Neural Accelerator in each GPU core](https://www.apple.com/newsroom/2026/08/apple-introduces-m6-and-m5-ultra-for-a-big-leap-in-performance-and-ai-compute/) 3. [Apple: iPhone 18 Pro tech specs, A20 Pro GPU with Neural Accelerators](https://www.apple.com/iphone-18-pro/specs/) 4. [Draw Things App Store listing and privacy label, version 26.1008.0](https://apps.apple.com/us/app/draw-things-ai-generation/id6444050820) 5. Draw Things for Mac binary, version 1.20260716.0: LC\_BUILD\_VERSION platform 6 (Mac Catalyst) 6. [Comfy Desktop 1.1.6 telemetry source (PostHog, Datadog RUM)](https://github.com/Comfy-Org/Comfy-Desktop/blob/9c43562d9e1e3a23af13f2e18653935e67bd9eda/src/main/lib/telemetry.ts#L69-L80) 7. [Comfy Desktop 1.1.6 first-launch screen: telemetry checked by default](https://github.com/Comfy-Org/Comfy-Desktop/blob/9c43562d9e1e3a23af13f2e18653935e67bd9eda/src/renderer/src/views/FirstUseTakeover.vue#L101) 8. [Comfy Desktop in-app privacy policy naming PostHog and Datadog](https://github.com/Comfy-Org/Comfy-Desktop/blob/9c43562d9e1e3a23af13f2e18653935e67bd9eda/src/renderer/src/lib/legalDocs.ts#L400-L401) 9. [Comfy docs: ComfyUI uses PyTorch MPS, not MLX](https://github.com/Comfy-Org/docs/blob/2c665405b557f2d8a36d75dfc41018ec31b18e56/installation/system_requirements.mdx#L69-L72) 10. Comfy Desktop 1.1.6 code signature: no App Sandbox entitlement