Skip to index

ComfyUI vs Draw Things

ComfyUI and Draw Things are both Image Generation tools in our directory. This side-by-side comparison covers pricing, editorial scores, supported platforms, key features and open-source status so you can decide which one fits your needs.

Side-by-side comparison

AttributeComfyUIDraw Things
Product statusStatus not recordedStatus not recorded
VendorComfy OrgDraw Things, Inc.
Rating★ 4.7★ 4.6
PricingFreemiumFreemium
PlatformsWindows, macOS, Linux, Self-hosted, Cloud, APIiOS, iPadOS, macOS
Key featuresNode-based workflow composition, JSON graph save/load, Model and custom-node configurationLocal image generation, Infinite canvas, On-device LoRA training
Headquarters——
Launched——
Open sourceYesNo

Key differences

  • ComfyUI carries the higher editorial score (4.7★ vs 4.6★).
  • Both ComfyUI and Draw Things use a Freemium model.
  • ComfyUI is open source (its code is public), while Draw Things is proprietary.
  • ComfyUI runs on Windows, macOS, Linux, Self-hosted, Cloud, API; Draw Things runs on iOS, iPadOS, macOS.

ComfyUI and Draw Things at a glance

ComfyUI

Comfy Org

Node-based media workflows with JSON graph exchange and custom extensions; local open-source software, paid cloud execution and external model calls have separate requirements and costs.

Rating
4.7★
Pricing
Freemium

Editorial verdict

Consider it when you need editable node workflows and can maintain their components. Confirm model permissions and local, cloud and external-call costs before production.

Strengths

  • Node-based composition and JSON workflow exchange
  • Local code available under GPL-3.0
  • Custom nodes and optional hosted/API workflows

Key features

  • Node-based workflow composition
  • JSON graph save/load
  • Model and custom-node configuration
  • Local software with optional paid cloud/Partner Nodes

Best for

  • Designing a versioned image or video workflow
  • Sharing a graph together with documented dependencies
  • Evaluating local versus hosted execution costs
Read the full ComfyUI review →

Draw Things

Draw Things, Inc.

Image generation and local LoRA training on iPhone, iPad and Mac, with a free local edition and optional account-based cloud compute.

Rating
4.6★
Pricing
Freemium

Editorial verdict

Consider the local edition for an Apple workflow you can configure. Confirm the chosen model license and whether each task runs locally or in the cloud.

Strengths

  • Free local workflow documented by the vendor
  • Canvas and LoRA training in the same app

Key features

  • Local image generation
  • Infinite canvas
  • On-device LoRA training
  • Optional cloud compute

Best for

  • Local image drafts on compatible Apple devices
  • LoRA experiments using material you are authorized to train on
Read the full Draw Things review →

Feature overlap

  • The two tools have no listed features in common — see both feature lists above.
  • ComfyUI also lists: Node-based workflow composition, JSON graph save/load, Model and custom-node configuration, Local software with optional paid cloud/Partner Nodes
  • Draw Things also lists: Local image generation, Infinite canvas, On-device LoRA training, Optional cloud compute

Frequently asked questions

Is ComfyUI or Draw Things better?

By editorial score, ComfyUI scores higher (4.7★ vs 4.6★). Both are capable Image Generation tools, so review the feature comparison for your use case.

Which is cheaper, ComfyUI or Draw Things?

Both ComfyUI and Draw Things use a Freemium pricing model.

Are ComfyUI and Draw Things open source?

ComfyUI is open source, while Draw Things is proprietary.

What platforms do ComfyUI and Draw Things support?

ComfyUI supports Windows, macOS, Linux, Self-hosted, Cloud, API; Draw Things supports iOS, iPadOS, macOS.

What are the main features of ComfyUI and Draw Things?

ComfyUI: Node-based workflow composition, JSON graph save/load, Model and custom-node configuration, Local software with optional paid cloud/Partner Nodes. Draw Things: Local image generation, Infinite canvas, On-device LoRA training, Optional cloud compute.

More Image Generation comparisons