.:🧩AOI :: Art Styles: A Visual Guide for AI Art :: ASCII Art: Images Made from Characters and Space

THE AOI STYLE ATLAS · EVERY IDEA LEADS SOMEWHERE

ASCII Art: Images Made from Characters and Space

Text-based visual art

Explore ASCII art through character choice, spacing, tonal ramps and native text. Learn to distinguish editable text artworks from raster interpretations.

AI-generated raster interpretation of a mountain valley and river constructed from pale character-like marks on charcoal; not a native ASCII text file.
AI-generated raster interpretation of a mountain valley and river constructed from pale character-like marks on charcoal; not a native ASCII text file.
Creative toolbox · take this style further

Choose an input and a destination. Search sends your words to the chosen site; Copy & open puts them on your clipboard to paste there.

Create with AI

Turn your chosen input into an image prompt or a new creative direction.

Image search

Find visual references. Short, concrete keywords usually work best.

Art & research

Explore artists, collections and the ideas behind a style.

Inspiration

Discover portfolios and visual ideas from creative communities.

A closer look

Visible rows and columns of characters; shapes made by punctuation and letters; whitespace used as light or negative space; density changes suggesting tone; a recognizable image that remains visibly constructed from text.

The characters are the material

Strict ASCII art uses printable characters from the ASCII character set, arranged with spaces and line breaks. RFC 20, published in 1969, documents seven-bit ASCII for network interchange; it is a technical reference for the character set, not a claim about who invented text-based art.

The name is also used more broadly in textmode communities. ASCII Gallery explains that some BBS-era works include extended DOS glyphs, while distinguishing plain text arrangements from ANSI works using terminal control sequences. State the convention being used. Unicode blocks, emoji and extended characters may be valid text art without being strict seven-bit ASCII.

Two useful ways to build an image

Contour-based work chooses characters for their shapes. A slash can suggest a rising edge, parentheses can turn a curve and a vertical bar can establish an upright. The arrangement is often designed by hand, with each character contributing to a recognizable outline or feature.

Tonal work uses character density to suggest light and dark. Sparse punctuation can create a pale area, while denser letters or symbols can form shadows. MIT Media Lab’s FaceTerm project discusses this familiar brightness-mapping approach and explores a more feature-sensitive use of typography for faces. Shape and tone can work together rather than requiring one universal conversion recipe.

A grid is a design constraint

Choose a fixed-width font and a manageable line width before drawing. Every character should occupy the expected column so the image does not collapse when displayed. Preserve spaces and line breaks, and avoid automatic wrapping. A text image that works in one narrow editor may need adjustment for another viewing environment.

Character cells are often taller than they are wide, so a square grid of characters may not produce a visually square image. Judge the result in its intended font. A circle that looks flattened or stretched can be corrected by changing the number of rows and columns rather than adding decorative detail.

Design for recognition first

Begin with a strong silhouette or a small number of distinctive features. A bird may need a clear beak and wing direction; a building may need a roofline and repeated openings. Leave enough empty space for those features to remain separate. Tiny gaps can matter as much as the printed marks.

For tonal images, test a short character ramp in the chosen font before using it across a whole picture. The apparent darkness of a character depends on its design and display. Preserve a hierarchy of large light and dark areas, then refine selected features. Converting every pixel mechanically can preserve noise while losing the subject.

Text files and raster interpretations

A native text artwork remains selectable characters with meaningful spacing. A screenshot or generated bitmap may resemble that appearance but does not provide an editable character arrangement. The new AI-generated gallery is labelled as raster interpretation: it illustrates textmode motifs without claiming every visible mark is a verified ASCII glyph or that the images are downloadable text files.

The prompt workshop therefore includes briefs for genuine text output as well as a visual interpretation. When making a native version, keep an original plain-text file, check its character set and preview it without proportional-font substitution. Credit creators when reusing existing text art; a small file or an old online circulation history does not remove authorship.

An exercise in deliberate character choice

Begin with a small object that has an unmistakable outline, such as a cup with a handle. Use a fixed-width editor and a frame no wider than forty columns. Draw only the rim, side, base and handle. Check whether the object reads before adding shading. Keep spaces as carefully as printed characters: closing the handle’s empty interior may destroy its identity.

Make a second version with a short tonal ramp chosen in the actual display font. Compare a space, punctuation and denser characters on the intended background; do not assume the same ordering works for every typeface or reversed colour scheme. Add only enough tone to distinguish the inside from the outside. If the shading overwhelms the contour, remove characters before increasing the canvas.

A portability check is part of finishing

Save a native text version, reopen it in a second fixed-width viewer and compare the longest lines, blank lines and leading spaces. Check that tabs have not shifted the arrangement and that automatic wrapping has not broken the silhouette. If a particular font, character encoding or terminal behaviour is essential, state that beside the work.

Present a short prose description alongside the text image so its subject is available without having to interpret every mark. For a raster version, supply an appropriate alt description and identify it as a raster. These are different deliverables: a visually convincing screenshot is not evidence that the underlying text will survive copying into another application.

Positive editorial assessment — AI-authored

ASCII art can achieve striking economy: a few characters and well-placed spaces can suggest form, gesture and mood. Its visible constraints invite inventive solutions. The strongest work remains readable both as an image and as an arrangement of individual typographic decisions.

Critical editorial assessment — AI-authored

Dense automatic conversions can become muddy, while font changes or wrapping can destroy carefully planned shapes. A raster imitation may conceal invalid glyphs. Improve the work by simplifying the subject, testing the actual display and preserving the native text when editability matters.

Neutral editorial assessment — AI-authored

ASCII art belongs to a wider field of text-based imagery. Strict character-set practice, extended textmode art and raster interpretations have different properties. Clear labelling helps viewers understand what they are seeing and how the work can be copied, displayed or modified.

A small poetic detour

Original words from the AOI studio.

Poetry

A slash leans into empty space;
a bracket turns to hold the air.
A row of dots becomes a distance,
a darker letter settles there.
The picture lives between the columns,
its silence measured line by line.
A little alphabet of choices
lets shape appear without a sign.

The prompt workshop

Original creative exercises. These prompts create new illustrations, not documentary evidence.

Native text: a small bird

Create an original plain-text ASCII drawing of a small bird in profile using only printable seven-bit ASCII characters and spaces. Maximum 40 columns and 18 rows. Preserve line breaks in a code block, use a fixed-width layout, and prioritize a readable beak, body and tail over dense shading.

Native text: a harbour

Create an original plain-text ASCII harbour scene using only printable seven-bit ASCII characters, spaces and line breaks. Maximum 72 columns and 28 rows. Include one sailboat, a simple quay and open sky. Use a coherent fixed-width grid and a restrained density ramp; output the artwork in a code block.

Raster concept: character portrait

Create an original raster illustration of a fictional adult face suggested entirely by visible rows of monospaced letters and punctuation, ivory marks on charcoal, strong light-dark hierarchy and generous margins. This is a visual textmode interpretation, not a claim of valid editable ASCII output.

A smile from the studio

Jokes

The portrait requested softer shading. The full stop offered to bring a few friends.

FOLLOW YOUR CURIOSITY

One style opens a thousand doors.

Find a familiar name, follow a visual idea, or put three styles side by side. Your next experiment starts here.

Collection snapshot · 2026-11-05 · Full database · Compare page templates