# Best open-source AI repos for AI image generation and editing (2026)

Source: https://dedupedai.com/best-ai-repos-for/image-generation/ (Deduped, updated 2026-10-26). Markdown twin of the page; JSON: https://dedupedai.com/best-ai-repos-for/image-generation/index.json

## What is the best open-source repo for AI image generation?

The best open-source AI repos for AI image generation and editing right now are ComfyUI, LocalAI and PhotoCraft. ComfyUI is the standard node-based tool for running image models locally; LocalAI serves image models next to LLMs behind one API; PhotoCraft is a Photoshop-style editor rather than a generator.

## Comparison

| Repo | What it does | Best for | Stars | Language |
|---|---|---|---|---|
| [ComfyUI](https://dedupedai.com/use-cases/comfy-org-comfyui/) (Comfy-Org/ComfyUI) | ComfyUI is an open-source node graph GUI, API, and backend for running diffusion models locally or in the cloud to generate images, video, audio, and 3D. | Creators, artists, and production teams who need full control over models and outputs and want to run them on their own hardware. | 141,884 | Python |
| [LocalAI](https://dedupedai.com/use-cases/mudler-localai/) (mudler/LocalAI) | LocalAI is an open-source AI engine that runs LLMs, vision, voice, image and video models on any hardware, with no GPU required. | Developers and teams who want a private, self-hosted replacement for cloud AI APIs that existing OpenAI or Anthropic clients can talk to. | 49,983 | Go |
| [PhotoCraft](https://dedupedai.com/use-cases/storytold-photocraft/) (storytold/photocraft) | PhotoCraft is an open-source, clean-room photo editor in pure Rust modeled on Photoshop: layers, masks, adjustment layers, layer styles, type, vectors, brushes and real PSD files, offline. | Designers and developers who want a free, offline, scriptable editor that opens PSDs, and people who want an AI agent to edit images. | 72,483 | Rust |
| [GPT Image Skill](https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/) (GENEXIS-AI/gpt-image-skill) | GPT Image Skill is an installable skill that lets Codex, Claude Code, or Google Antigravity generate and edit GPT images through a ChatGPT subscription instead of the Images API. | Designers, marketers, and developers who already pay for ChatGPT and want image generation inside their coding agent without setting up API keys. | 180 | JavaScript |
| [Awesome Qwen Image](https://dedupedai.com/use-cases/wildminder-awesome-qwen-image/) (wildminder/awesome-qwen-image) | A curated list of checkpoints, quantizations, prompt rewriters, LoRAs, and tooling for Qwen-Image 2.1, Alibaba's 7B unified text-to-image and image-editing model. | Creators, developers, and AI agents who need to pick the right Qwen-Image 2.1 file for their runtime, VRAM budget, or Mac. | 157 | n/a |

## When to pick which

### ComfyUI (Comfy-Org/ComfyUI)

ComfyUI is an open-source node graph GUI, API, and backend for running diffusion models locally or in the cloud to generate images, video, audio, and 3D.

Best for: Creators, artists, and production teams who need full control over models and outputs and want to run them on their own hardware.

Needs: A Windows or macOS computer for the desktop app, or a Windows, Linux, or macOS machine with a supported GPU for the manual install; A GPU with enough VRAM for the model you pick (the README says large models can run on as little as 4GB VRAM with 8GB RAM); Disk space for the model files you choose to download; Optional: a Comfy Cloud account or partner node access for paid hosted models.

Start here: [Generate images from ready-made template workflows](https://dedupedai.com/use-cases/comfy-org-comfyui/generate-images-with-template-workflows/) (Beginner, 30 minutes)

1. Download the Desktop Application from https://www.comfy.org/download (available on Windows and macOS).
2. Open a template from https://comfy.org/workflows and load it in ComfyUI.
3. Change the prompt text in the text input node and set the model loader to a model you have installed.
4. Press Ctrl + Enter to queue the current graph for generation.

All 4 use cases: https://dedupedai.com/use-cases/comfy-org-comfyui/ | GitHub: https://github.com/Comfy-Org/ComfyUI

### LocalAI (mudler/LocalAI)

LocalAI is an open-source AI engine that runs LLMs, vision, voice, image and video models on any hardware, with no GPU required. Each backend (llama.cpp, vLLM, whisper.cpp, stable-diffusion, MLX and more) is pulled only when a model needs it, and every backend sits behind drop-in OpenAI, Anthropic and ElevenLabs compatible APIs, with multi-user auth, quotas and built-in agents with tools, RAG and MCP.

Best for: Developers and teams who want a private, self-hosted replacement for cloud AI APIs that existing OpenAI or Anthropic clients can talk to.

Needs: Docker or Podman, or the macOS app (not signed by Apple); Optional GPU: NVIDIA, AMD, Intel, Apple Silicon or Vulkan; CPU-only works; Disk space for the models you pull.

Start here: [Self-host an OpenAI-compatible API on your own server](https://dedupedai.com/use-cases/mudler-localai/self-host-an-openai-compatible-api/) (Beginner, 30 minutes)

1. Start the CPU image: docker run -ti --name local-ai -p 8080:8080 localai/localai:latest
2. Or with an NVIDIA GPU: docker run -ti --name local-ai -p 8080:8080 --gpus all localai/localai:latest-gpu-nvidia-cuda-12
3. Load a model, for example local-ai run llama-3.2-1b-instruct:q4_k_m
4. Point your app's OpenAI base URL at http://localhost:8080

All 3 use cases: https://dedupedai.com/use-cases/mudler-localai/ | GitHub: https://github.com/mudler/LocalAI

### PhotoCraft (storytold/photocraft)

PhotoCraft is an open-source, clean-room photo editor in pure Rust modeled on Photoshop: layers, masks, adjustment layers, layer styles, type, vectors, brushes and real PSD files, offline. Every menu item and tool runs one of 500+ commands, so the same engine can be driven from the UI, a headless CLI, a JSON control channel or an MCP server. The README says it is early alpha and not yet a Photoshop replacement for daily professional work.

Best for: Designers and developers who want a free, offline, scriptable editor that opens PSDs, and people who want an AI agent to edit images.

Needs: A download from the PhotoCraft page, or Rust to build from source.

Start here: [Open, edit and save PSD files without Photoshop](https://dedupedai.com/use-cases/storytold-photocraft/edit-psd-files-without-photoshop/) (Beginner, 10 minutes)

1. Download PhotoCraft from its page on getartcraft.com, or build it from source with Rust (cargo run --release -p PhotoCraft)
2. Open a PSD; menus, panels and shortcuts follow Photoshop
3. Edit layers, masks, adjustment layers and type
4. Save as PSD or export to PNG, JPEG, TIFF or WebP

All 3 use cases: https://dedupedai.com/use-cases/storytold-photocraft/ | GitHub: https://github.com/storytold/photocraft

### GPT Image Skill (GENEXIS-AI/gpt-image-skill)

GPT Image Skill is an installable skill that lets Codex, Claude Code, or Google Antigravity generate and edit GPT images through a ChatGPT subscription instead of the Images API.

Best for: Designers, marketers, and developers who already pay for ChatGPT and want image generation inside their coding agent without setting up API keys.

Needs: A ChatGPT or Codex plan that permits image generation (the README says Codex image generation is not available on the Free plan); Node.js 22 or newer; Git, for installing from GitHub; Codex CLI signed in with ChatGPT, or a host that provides subscription-native image_gen.

Start here: [Generate images from a text prompt in Codex](https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/generate-images-from-text-prompt/) (Intermediate, 30 minutes)

1. Paste the install prompt from the README into Codex, Claude Code, or Google Antigravity, and follow the Sign in with ChatGPT device authorization when asked.
2. Wait for the setup to report readiness. The README states that no image is generated during setup.
3. In Codex, type $gpt-image Create a cozy reading room at sunset, 16:9, high quality. In Claude Code, use /gpt-image with the same request.
4. Open the returned PATH in /generated-images and check the image against your request.

All 5 use cases: https://dedupedai.com/use-cases/genexis-ai-gpt-image-skill/ | GitHub: https://github.com/GENEXIS-AI/gpt-image-skill

### Awesome Qwen Image (wildminder/awesome-qwen-image)

A curated list of checkpoints, quantizations, prompt rewriters, LoRAs, and tooling for Qwen-Image 2.1, Alibaba's 7B unified text-to-image and image-editing model.

Best for: Creators, developers, and AI agents who need to pick the right Qwen-Image 2.1 file for their runtime, VRAM budget, or Mac.

Needs: A runtime that matches the chosen entry point: Diffusers, ComfyUI, llama.cpp, or MLX on Apple Silicon; Enough disk space: the base text encoder alone is about 17.5 GB in BF16; A GPU with 8 to 12 GB VRAM for the INT4 ConvRot pack, or a Mac with Apple Silicon for MLX-4bit; Acceptance of the Qwen Research License that covers the base weights.

Start here: [Generate images with the official Qwen-Image-2.1 base model](https://dedupedai.com/use-cases/wildminder-awesome-qwen-image/generate-images-with-official-base-model/) (Intermediate, 1 hour)

1. Open the official Qwen-Image-2.1 model card on Hugging Face and read the license and aspect-ratio table.
2. Download the whole Qwen/Qwen-Image-2.1 repo, not a single file, because the DiT and text encoder ship as numbered safetensors shards with an index.
3. Load the model with the QwenImage21Pipeline class in Diffusers.
4. Run a sample prompt using the default of 40 steps without classifier-free guidance, then try CFG if prompt adherence is weak.

All 5 use cases: https://dedupedai.com/use-cases/wildminder-awesome-qwen-image/ | GitHub: https://github.com/wildminder/awesome-qwen-image

## Questions

### What is the best open-source repo for AI image generation?

The best open-source AI repos for AI image generation and editing right now are ComfyUI, LocalAI and PhotoCraft. ComfyUI is the standard node-based tool for running image models locally; LocalAI serves image models next to LLMs behind one API; PhotoCraft is a Photoshop-style editor rather than a generator.

### Which repo for AI image generation and editing is easiest to start with?

ComfyUI: its starter use case "Generate images from ready-made template workflows" is rated Beginner and takes about 30 minutes.

### How do I get started with ComfyUI?

Download the Desktop Application from https://www.comfy.org/download (available on Windows and macOS); Open a template from https://comfy.org/workflows and load it in ComfyUI; Change the prompt text in the text input node and set the model loader to a model you have installed. Full steps and a copy-paste agent prompt: https://dedupedai.com/use-cases/comfy-org-comfyui/generate-images-with-template-workflows/

### Are these AI image generation and editing repos free?

They are open-source projects on GitHub (check each license; some, like fair-code projects, limit commercial use). Some also sell a hosted version or need a paid model API key; the requirements for each repo are listed on this page.

### How is this list of AI image generation and editing repos chosen?

Deduped tracks the hottest AI repos on GitHub and writes step-by-step use cases from each README. For this page we picked the repos that do this job, ordered by fit and then GitHub stars. Repos we only inferred from a short README, or that we do not index, are left out. The page is rebuilt with the site; stars are as of the last build in 2026.

Third-party repos we don't own or vouch for. AI-written use cases can be wrong: review everything and run at your own risk.