# Open-source alternatives to Midjourney

> We track 6 open-source alternatives to Midjourney, but none currently has confirmed commercial use, local running and no paid services together. ComfyUI, Fooocus, Krita AI Diffusion and SD.Next cover all 4 Midjourney jobs.

Midjourney: Paid hosted text-to-image service with image editing, upscaling and style references. Pricing: See pricing page (checked 2026-10-15). Facts checked 2026-10-15.

## Projects

| Project | Jobs | Runs locally | Commercial use | Paid services | Hardware |
|---|---|---|---|---|---|
| [ComfyUI](https://repoquarry.dev/projects/comfyui) | text-to-image, image-editing, upscaling, style-reference, songs-with-vocals | Yes, fully offline; optional paid Comfy Cloud | GPL-3.0 code; weights vary (e.g. FLUX.1-dev is non-commercial) | None required; paid Comfy API nodes and Comfy Cloud are optional | Big models from 4GB VRAM + 8GB RAM; CPU-only portable build exists |
| [InvokeAI](https://repoquarry.dev/projects/invokeai) | text-to-image, image-editing, upscaling | Yes, local web server and UI | Apache-2.0 code; weights vary (SD3.5 community licence, FLUX.1-dev non-commercial) | None required; some models are API-only (GPT Image, Nano Banana, Wan) | SD1.5 4GB VRAM/8GB RAM; SDXL 8GB/16GB; FLUX.1 10GB/32GB |
| [Krita AI Diffusion](https://repoquarry.dev/projects/krita-ai-diffusion) | text-to-image, image-editing, upscaling, style-reference | Yes, local ComfyUI server; optional paid cloud | GPL-3.0 code; weights vary (Flux 2 Klein 4B and Z-Image Turbo are Apache-2.0) | None required; optional cloud tokens (10 EUR for 5,000) | 6GB NVIDIA VRAM recommended; CPU supported but very slow |
| [SD.Next](https://repoquarry.dev/projects/sdnext) | text-to-image, image-editing, upscaling, style-reference | Yes, local server and web UI; Docker recipes | Apache-2.0 code; weights vary (e.g. FLUX.1-dev is non-commercial) | None required | CPU-only supported; SDNQ cuts VRAM up to 4x; no minimums stated |
| [Stable Diffusion web UI (AUTOMATIC1111)](https://repoquarry.dev/projects/stable-diffusion-webui) | text-to-image, image-editing, upscaling | Yes, local web UI | AGPL-3.0 code; SD 1.5 OpenRAIL-M allows commercial use with restrictions | None required | 4GB GPU supported (reports of 2GB); NVIDIA recommended |
| [Fooocus](https://repoquarry.dev/projects/fooocus) | text-to-image, image-editing, upscaling, style-reference | Yes, offline | GPL-3.0 code + OpenRAIL-M weights; README calls it non-commercial software | None required | 4GB NVIDIA VRAM + 8GB RAM; CPU-only needs 32GB RAM (~17x slower) |

## Common questions

### Is there a free, open-source alternative to Midjourney?

Yes. ComfyUI, SD.Next, Krita AI Diffusion, InvokeAI and Fooocus are free to download and run on your own computer without paid API keys. ComfyUI, SD.Next, Krita AI Diffusion and Fooocus cover every job we list for Midjourney; the other covers only part of it, so check which job you need.

### Can I do text-to-image locally without a Midjourney subscription?

Yes, with ComfyUI (GPL-3.0), InvokeAI (Apache-2.0) and SD.Next (Apache-2.0). Check the hardware notes and licence conditions on each page before you start.

### Can I do image editing (inpainting and outpainting) locally without a Midjourney subscription?

Yes, with ComfyUI (GPL-3.0), InvokeAI (Apache-2.0) and SD.Next (Apache-2.0). Check the hardware notes and licence conditions on each page before you start.

### Which Midjourney alternatives can I use commercially?

With conditions: ComfyUI, Fooocus, InvokeAI, Krita AI Diffusion, SD.Next and Stable Diffusion web UI (AUTOMATIC1111). This is not legal advice; read each licence.

### Does any open-source project replace all of Midjourney?

ComfyUI, Fooocus, Krita AI Diffusion and SD.Next cover every job we list for Midjourney.

---
Source: RepoQuarry (https://repoquarry.dev/alternatives/midjourney). Data licence: CC BY 4.0. Quotes are from project documentation and belong to their authors.