AiCadGen
The core workflow is prompt-in, parametric-geometry-out: the vendor states the AI writes parametric code, runs it in a sandbox, and…
Design AI tools cover a widening surface area: UI mockups from a prompt, brand kits from a one-page brief, presentations that assemble themselves, 3D assets for games and product visualization, and illustrations that match a defined style. The category has splintered because design itself splinters — the person building a pitch deck and the person laying out a mobile app have different jobs, different reference materials, and different tolerances for AI weirdness. Pick based on the specific deliverable you need to ship, not the general promise of "AI design."
The core workflow is prompt-in, parametric-geometry-out: the vendor states the AI writes parametric code, runs it in a sandbox, and…
The output is a single high-res PNG containing front, 3/4, side, and back views alongside expressions, action poses, a hex-coded color…
Framesmith is an MIT-licensed MCP server for creating and reviewing interface designs with an AI assistant.
DryRunn simulates the rehearsal environment for board presentations, investor pitches, sales strategy reviews, and executive Q&A sessions…
The core loop is prompt or image in, 3D model out, then refine — texture, remesh, repair, convert — before exporting into whichever format…
AiPPT accepts PDFs, TXT files, Word documents, Google Docs, Google Slides, and webpage URLs as source material, then generates a full…
GenUI is a native Swift workspace where agents produce declarative messages describing UI components, the client validates those messages…
Figviz takes a plain-English prompt or an attached PDF and returns a labeled, science-accurate diagram sized for slides, worksheets, or…
The workflow is conversational: describe the layout, the sections, the feel, and PixelGlass writes Handlebars templates with a live preview…
Alai takes a text prompt and generates a full presentation, handling layout, visual hierarchy, and slide-to-slide design coherence without…
Framer is a design-to-publish tool that now pairs its visual editor with AI features for generating sections, content, and full landing pages from a prompt. It is the right pick when the output is the live site, not a static mockup handed to a developer.
Beautiful.ai applies opinionated design rules to every slide you add, with AI that generates decks from an outline and reflows content when you edit. It consistently produces presentations that do not embarrass you in a boardroom.
AIdea is a lightweight generator for presentation decks and creative briefs. It is useful when you need a credible draft in minutes and are willing to polish the output in a more capable editor afterward.
The design directory on this site is still growing into its leaf categories. UI generators (Uizard, Galileo, v0), 3D-asset tools (Luma, Meshy), brand-kit generators, and the rapidly evolving space of AI-native moodboarding tools will appear here as listings are verified. Until then, pair one of the tools above with a proper design editor (Figma, Adobe Creative Cloud, Rive) rather than trying to make a single AI tool do the whole job.
Not for anything where the brand or the product has to feel considered. AI compresses the speed of first drafts and option exploration, which is where designers already spend a lot of time. The judgment layer — what to ship, why, to whom — is still a human job.
On paid tiers of the major tools, generally yes, with the usual caveats around trademarked logos and celebrity likenesses. Read the terms on the specific tool before a production launch.
For teams already on Figma, plugins reduce context switching and keep everything in the file your team already knows. Standalone tools make sense when the deliverable is a full website, a deck, or a 3D asset that Figma does not produce.
Generate several variations up front, narrow to two or three, then iterate on the chosen one inside a tool that supports comments and version history. Doing the narrowing in the AI tool and the polish in a proper design editor is a common and effective pattern.
AI design tools that do not respect your design system are counterproductive. Before adopting a tool at a team level, test whether it can consume and produce outputs aligned with your component library, tokens, and typography scale. If it cannot, limit its use to ideation and keep production design in a system-aware environment.
Write the brief the way you would for a human designer: the audience, the goal, the constraints (size, aspect ratio, brand rules), the tone, and references you admire. Vague prompts produce vague output, and the cost of a thorough brief is almost always less than the cost of regenerating until you get lucky.