HappyAccidents
makes it
that simple to generate art using A.I.
Working with A.I. models is hard...
That is why we built HappyAccidents to manage models, embeddings, datasets, etc for you in the cloud. So you can spend all your time creating.
"Times Square, New York City in the style of Studio Ghibli" - using Studio Ghibli Diffusion
Generate images from any open-source model
If a model is available on HuggingSpace or Civitai, you can use it seamlessly, without manual downloading, uploading, etc.
"Matte concept art of city built alongside a river. Lush, green trees in the foreground." - Using Img2Img
Create with ease
Whether it's designing a board game, creating a video game, editing your photos, or any other passion project, let us help you bring it to life.
Made with HappyAccidents
Train your own embeddings
Teach the A.I about certain people, things, styles, etc, just from a few photos. Use your subjects and styles to create anything you can imagine.
Powerful but simple
Our goal for Happy Accidents is to provide all the core features of Automatic1111, in a user interface so simple that even your parents could use it.
Generate with any CivitAI model
Use any open-sourced model, LoRA, Textual Inversion from CivitAI without the hassle.
Leverage multiple ControlNets
All the latest ControlNets to accomplish anything you're working on
Inpainting
Iterate on your work till it's perfect. We support drawing custom masks.
Upscaling
We support the common Upscalers (ESRGAN, Karras, etc)! We'll support any we don't have on request.
Face Restore
We currently enable GFPGAN & CodeFormer, and we can easily add more on request.
Custom training
Coming SoonTake any open source model and train it with your own custom styles and subjects.
Start creating for free
Get access to all the models, LoRAs, and Textual Inversions on CivitAI, Inpainting, and Upscaling. All for free.