# The Draw Things alternative for people who want the picture. You wanted a picture. You got a pipeline. Draw Things gives you 248 models, 20 samplers, and a setting for nearly everything. That's a lot of power, and a lot of decisions before the first image. Private Diffusion ships a curated set of models, each trained to hold its quality, and runs every one your iPhone, iPad, or Mac has the memory for. 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 Draw Things Picture quality Private DiffusionRetrained after shrinking, so the small model answers like the original Draw ThingsCompressed after training, shipped as is Speed on Apple Silicon Private DiffusionGPU and Neural Engine, every model Draw ThingsGPU; Neural Engine only for select models What leaves your device Private DiffusionCollects no data, not even crash reports Draw ThingsPurchases, User Content, Identifiers, per its App Store label The app itself Private DiffusionNative Swift on iPhone, iPad, and Mac, with Liquid Glass Draw ThingsA Mac Catalyst build of its iPad 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 To fit an image model on a phone, you have to shrink its weights. Draw Things does that after training: it compresses the full-precision weights and ships them without retraining, and its default Krea 2 download is an 8-bit file. We take one more step. 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. The training comes after the compression, so it gets to repair what the compression bent. 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. Draw Things Draw Things compresses the full-precision weights after training and ships them without retraining. ## Faster, Because Both Processors Work Private Diffusion puts the GPU and the Neural Engine to work together, and you never have to find a setting to make that happen. How it works The diffusion model runs on the GPU through MLX, Apple's own machine learning framework, and it 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. On iPhone 18 Pro and Macs with Ultra chips, large images decode across both Neural Engines. Draw Things uses the Neural Engine too, but by default only for its 8-bit S models, and only on some chips. Its default Krea 2 download never touches it. Our 4-bit weights also move half the bytes per step of that 8-bit default, and image generation spends most of its time waiting on memory, so fewer bytes means less waiting. Sources[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/)[Apple: iPhone 18 Pro tech specs, A20 Pro GPU with Neural Accelerators](https://www.apple.com/iphone-18-pro/specs/) Which Processors Do the Work Private Diffusion - GPUUsed - Neural EngineUsed Draw Things - GPUUsed - Neural EngineSelect models only ## Nothing Collected, Not Even Your Purchases Private Diffusion collects no data at all, while the App Store label for Draw Things lists Purchases, User Content, and Identifiers. How it works Every App Store app has to declare what it collects. The [privacy label for Draw Things](https://apps.apple.com/us/app/draw-things-ai-generation/id6444050820) lists Purchases, User Content, and Identifiers as data it may collect, not linked to your identity. Its optional Cloud Compute also runs generations on its servers once you sign in. Private Diffusion collects nothing. It never asks for an account, ships without an analytics SDK or a crash reporter, and makes every image on your device. If you're signed in to iCloud, your gallery syncs to your own private iCloud, never to a server of ours. Sources[Draw Things App Store listing and privacy label, version 26.1008.0](https://apps.apple.com/us/app/draw-things-ai-generation/id6444050820) ![Draw Things' App Store privacy nutrition label, listing Purchases, User Content, and Identifiers as data collected but not linked to your identity.](/_astro/draw-things-app-privacy-card.BgEaSrwX_1Ge4YQ.png) Draw Things' own App Store privacy label, version 26.1008.0, checked October 13, 2026. Draw Things App Store listing and privacy label, version 26.1008.0 ## A Real Mac App, Built the Way Apple Builds Them Private Diffusion is a native Swift app with Liquid Glass on iPhone, iPad, and Mac, built to Apple's Human Interface Guidelines. How it works On the Mac, Draw Things is a dated Mac Catalyst app: a Catalyst build of its iPad app, with the iPad interface carried over to the desktop. Private Diffusion is written in Swift for macOS 26.4 and later, with Liquid Glass, menu bar commands, keyboard shortcuts, and the App Sandbox and hardened runtime switched on. It follows Apple's Human Interface Guidelines to the T, on every device it runs on. SourcesDraw Things for Mac binary, version 1.20260716.0: LC\_BUILD\_VERSION platform 6 (Mac Catalyst) ![Draw Things for Mac: a sidebar listing Projects, Settings, Control, Scripts, and PEFT, a checkerboard canvas, a Basic Settings column with Model, LoRA, Control, and Strength, and a Tools column.](/_astro/draw-things-mac.K8r7XvP8_Z1h8VmX.png) Draw Things for Mac, version 1.20260716.0, the iPad interface in a Catalyst window. Checked October 13, 2026. Draw Things for Mac binary, version 1.20260716.0: LC\_BUILD\_VERSION platform 6 (Mac Catalyst) ![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 Draw Things If - You want 248 models and every sampler within reach - You train your own LoRAs - You want a setting for every stage of the pipeline ### Pick Private Diffusion If - You want the picture, not the pipeline - You want an app that collects nothing, not even purchases - You want a native Mac, iPhone, and iPad app that follows Apple's guidelines ## Private Diffusion and Draw Things Side by Side How the two apps compare on Apple Silicon, as of the date in the sources below. | What you get | Private Diffusion | Draw Things | | --- | --- | --- | | How models are shrunk | Retrained after shrinking (QAD) | Compressed after training, no retraining | | Processors it uses | GPU and Neural Engine | GPU; Neural Engine for select models | | Data the app collects | None | Purchases, User Content, Identifiers | | Analytics SDKs | None | None named in its policy | | How it's built | Native Swift with Liquid Glass | Mac Catalyst build of its iPad app | | App Sandbox | Yes, with hardened runtime | Yes | | Runs on | iPhone, iPad, and Mac | iPhone, iPad, and Mac | | Getting started | Download a model, type a prompt | Pick a model, then tune the settings | | Models | Curated, each tuned for your device | 248 catalog entries | 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 | ## Draw Things Alternative FAQ - ### What is the best Draw Things alternative for Mac? For people who want an image rather than a workflow, Private Diffusion. It runs a curated set of models on Apple Silicon Macs with macOS 26.4 or later, retrains each model after shrinking it so the quality holds, uses the GPU and the Neural Engine together, and collects no data. Draw Things is still the better pick if you want hundreds of models and control over every setting. - ### Is Draw Things a native Mac app? Draw Things for Mac is a Mac Catalyst build of its iPad app, so it brings the iPad interface to the desktop. It is sandboxed and runs on macOS 12.4 or later. Private Diffusion is a native Swift app for macOS 26.4 and later, with Liquid Glass, menu bar commands, and keyboard shortcuts. - ### Can I use Draw Things on iPhone and iPad? Yes. Draw Things runs on iPhone, iPad, and Mac, and so does Private Diffusion. The difference is what happens after you open it: Draw Things asks you to pick from 248 catalog entries and tune the settings, while Private Diffusion gives you models already matched to your device. 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. - ### Is Draw Things safe? Draw Things is a legitimate app from Draw Things, Inc., sandboxed on the Mac and sold through the App Store. Its App Store privacy label says it may collect Purchases, User Content, and Identifiers for app functionality, not linked to your identity, and its optional Cloud Compute runs generations on its servers after you sign in. Private Diffusion collects no data, and every image is made on your device. ## 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 ComfyUI](/en/comfyui-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