Stable Diffusion
Stable Diffusion is a family of latent diffusion models that generates and edits images from text and other conditioning inputs.
Build a Text-to-Image Generator (from Scratch)
Stable Diffusion is a latent diffusion model for generating and editing images from text and other conditioning inputs.
- How it is used
- Generates and edits images from text and other conditioning inputs through accessible latent-diffusion tooling.
- How deep it goes
- Explains the three-component architecture, tensor shapes, reverse diffusion, inference steps, and practical pipeline parameters.
- Examples
- Uses prompts about astronauts, unicorns, animals, beaches, lions, and tigers to demonstrate generation and embedding interpolation.
- What you can do after
- Apply it via Hugging Face Diffusers to generate images, vary inference settings, and blend text embeddings into hybrid outputs.
Read how Build a Text-to-Image Generator (from Scratch) explains Stable Diffusion →
Generative AI in Action
Stable Diffusion is a family of diffusion-based generative models that produces images from textual and other conditioning inputs.
- How it is used
- Supports enterprise image generation and editing across architecture, fashion, e-commerce, entertainment, and healthcare applications.
- How deep it goes
- Explains the text encoder, UNet, scheduler, and VAE, then demonstrates API parameters without deriving the underlying mathematics.
- Examples
- Demonstrates a laughing panda prompt, an SDXL REST request, API header and body parameters, and the resulting generated image.
- What you can do after
- Apply it via a Stability REST API using an engine ID, bearer authentication, prompts, dimensions, weights, and generation controls.
Read how Generative AI in Action explains Stable Diffusion →
Deep Learning with JAX
A widely used pretrained text-to-image diffusion model.
- How it is used
- Generates images from text prompts through a pretrained diffusion pipeline.
- How deep it goes
- Explains text encoding and multistep diffusion, but defers the model’s internal mathematics.
- Examples
- Uses the runwayml/stable-diffusion-v1-5 checkpoint, a Flax pipeline, bfloat16 parameters, TPU execution, and eight generated images.
- What you can do after
- Apply it via Diffusers to generate images and assemble custom diffusion pipelines.
Related concepts
- image generation modelkind of · An image generation model is a machine-learning model that creates, edits, or transforms images from learned patterns and conditioning inputs.
- tokenuses · A token is a discrete unit of text, such as a word, word fragment, character, or symbol, represented and processed as an element by a language model.
- diffusion modelkind of · A diffusion model is a generative model that learns to produce data by reversing a process that progressively adds noise to data.
- DALL-Ealternative to · DALL-E is a family of OpenAI generative models that creates and edits images from natural-language descriptions and other conditioning inputs.
- digital imageproduces · A digital image is visual content encoded as data, typically as a two-dimensional arrangement of pixels or equivalent visual elements.
- Midjourneyalternative to · Midjourney is a generative artificial-intelligence service that creates images from natural-language prompts and other user-provided inputs.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.