| S Stable Diffusion | Free & Open Source | A leading open-source image generation model that runs locally or via hosted services, with a large community of models and tools. | - Fully open source under Apache 2.0 — run locally with no usage fees or rate limits
- Huge ecosystem of fine-tuned models, LoRAs, and community tools on Civitai and HuggingFace
| - Local setup requires a capable GPU and technical knowledge to configure
- Image quality varies significantly depending on the model version and prompt skill
| Visit → |
| | Free & Open Source FLUX.1 Schnell is free under Apache 2.0; FLUX.1 Dev requires a non-commercial licence; hosted API tiers available via BFL. | State-of-the-art open-source image generation models by Black Forest Labs, with FLUX.1 Schnell freely available under Apache 2.0. | - FLUX.1 Schnell produces photorealistic results that rival Midjourney's quality
- Open source and commercially usable under Apache 2.0 for the Schnell variant
| - Best results require a capable GPU or a paid hosted API endpoint
- Lacks Midjourney's simple web interface and community image gallery
| Visit → |
| A Adobe Firefly | Free Free tier with monthly generative credits; additional credits or unlimited generation included in Creative Cloud subscriptions. | Adobe's commercially safe AI image generator, trained on licensed content and available via web, Photoshop, and Express. | - Commercially safe — trained on licensed content, reducing copyright risk for businesses
- Integrates directly into Adobe Photoshop, Illustrator, and Express workflows
| - Not open source; tied to Adobe's ecosystem and subscription pricing model
- Free tier monthly credit allowance runs out quickly with heavy generation use
| Visit → |
| L Leonardo.ai | Free Free tier with 150 daily tokens that reset every day; paid plans from $10/month for more tokens and priority generation. | A web-based AI art platform with fine-tuned models, image-to-image tools, and a daily free token allowance for image generation. | - Daily free token allowance resets every 24 hours — genuinely usable at no cost
- Large selection of community-trained and fine-tuned models for different art styles
| - Not open source; proprietary platform with all generation done on cloud servers
- Daily token limit is low enough to frustrate users generating images at volume
| Visit → |