DALL-E
DALL-E is a family of OpenAI generative models that creates and edits images from natural-language descriptions and other conditioning inputs.
Data Storytelling with Altair and AI
A generative AI system that creates and edits images from textual descriptions or image-editing instructions.
- How it is used
- Adds generated visual context, annotations, and call-to-action icons to data stories.
- How deep it goes
- Defines text-to-image prompting and demonstrates generation, selection, editing, remote hosting, and Altair placement; defers advanced image techniques.
- Examples
- Uses athlete cartoons, homelessness photos, animal images, and black-and-white customer-experience and dollar-bag icons.
- What you can do after
- Apply it to create, select, refine, and place prompt-generated images in Altair data stories.
Read how Data Storytelling with Altair and AI explains DALL-E →
Build a Text-to-Image Generator (from Scratch)
A text-to-image generative model that produces images conditioned on natural-language descriptions.
- How it is used
- Generates images from natural-language prompts through transformer-based text encoding, image-token prediction, and visual reconstruction.
- How deep it goes
- Traces training and inference through BART and VQGAN components, including tokenization, prediction, reconstruction, and cross-entropy supervision.
- Examples
- Uses paired image–caption datasets and the prompt “panda with top hat reading a book” with staged token visualizations.
- What you can do after
- Build it from scratch with PyTorch by implementing text encoding, autoregressive image-token prediction, and VQGAN reconstruction.
Read how Build a Text-to-Image Generator (from Scratch) explains DALL-E →
Build a Website with ChatGPT
DALL-E is a generative model that creates images from textual descriptions.
- How it is used
- Supports creating visual assets for online journal pages, including logos, icons, illustrations, and other page imagery.
- How deep it goes
- Explains text-to-image prompting, access options, costs, and limitations, but stops short of model internals.
- Examples
- Demonstrates cartoons, photographs, digital art, and illuminated manuscript pages through concrete prompts.
- What you can do after
- Apply it via the OpenAI DALL-E app or Copilot Image Creator to generate image drafts from detailed prompts.
Learn Generative AI with PyTorch
A text-to-image generative model that produces images from textual descriptions.
- How it is used
- Generates images from textual descriptions through text conditioning and diffusion-based denoising.
- How deep it goes
- Explains text encoding, CLIP alignment, and iterative denoising, while using the diffusion model’s internals as the deeper foundation.
- Examples
- Uses astronaut-and-unicorn and cat-at-a-computer prompts with a Python image-generation call.
- What you can do after
- Apply it via the OpenAI Python client to generate and retrieve an image from a text prompt.
Related concepts
- digital imageproduces · A digital image is visual content encoded as data, typically as a two-dimensional arrangement of pixels or equivalent visual elements.
- 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.
- transformer-based text-to-image modelkind of · A transformer-based text-to-image model generates images from text by modeling relationships between textual and visual token sequences with transformer architectures.
- promptuses · An input, typically containing instructions or context, supplied to an artificial-intelligence model to guide or elicit its output.
- Stable Diffusionalternative to · Stable Diffusion is a family of latent diffusion models that generates and edits images from text and other conditioning inputs.
- CLIPuses · CLIP is a multimodal model and pretraining method that aligns image and text representations in a shared embedding space.
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.