AI Character Generator: Original Characters for D&D, Novels and Games
Describe a character, species to boots, and get full-figure concept art on a plain backdrop.
Updated
IllustrationExamples
Illustrations of the kind of result this makes. They are not app outputs.
Illustration
Illustration
Illustration
Illustration
IllustrationHow it works
Describe the character
Species or age and build, face, outfit and gear, palette, pose, then the art style and a plain backdrop. One character per image.
Draft, then finalise
Run it on Flux Schnell for 3 gems until the design reads right, then run the same wording on FLUX 1.1 Pro for 18.
Keep a character sheet
Save the exact prompt and the images you like. The same words plus a new pose line give the closest match in later runs.
Which model for which kind of character art?
| Model | Cost per image | Input | Pick it for |
|---|---|---|---|
| Flux Schnell | 3Gems | Text | First drafts of a design; trying several outfits, palettes or styles cheaply |
| FLUX 1.1 Pro | 18GemsPaid plan | Text | The final character sheet image in 4:5, 9:16 or 3:2 with the most detail |
| Nano Banana | 12Gems | Your photo plus text | Turning a photo of yourself into a character while keeping your face |
The cost is shown next to the model before you generate, and a failed generation returns its gems. Plans and gem packs are on the pricing page.
Character briefs to copy or run
IllustrationIllustration made from this prompt with an image model, not a result from the UniqAI app.
Full-figure concept art of an original fantasy ranger: a tall woman in her thirties with dark brown skin and close-cropped hair, weathered green cloak, scaled leather armour, a longbow on her back, standing relaxed. Painterly digital illustration, soft overcast light, muted forest palette, plain warm grey backdrop, vertical frame.
Run this briefIllustrationIllustration made from this prompt with an image model, not a result from the UniqAI app.
Full-figure concept art of a starship pilot: a stocky man in his forties with a grey beard and a black prosthetic left hand, scuffed orange and white pressure suit with worn patches, helmet under his arm, standing square to camera. Clean sci-fi illustration, cool studio light, plain pale grey backdrop, vertical frame.
Run this briefIllustrationIllustration made from this prompt with an image model, not a result from the UniqAI app.
Full-figure character design of a kindly tea-shop owner: a round older woman with silver curls, round glasses, a mustard cardigan and a flour-dusted apron, holding a steaming teapot. Warm storybook illustration with soft gouache texture, golden afternoon light, plain cream backdrop, vertical frame.
Run this briefIllustrationIllustration made from this prompt with an image model, not a result from the UniqAI app.
Full-figure concept art of a steampunk inventor: a wiry person in their fifties with goggles pushed up into wild grey hair, a brass mechanical left arm, a leather waistcoat lined with tools and ink-stained fingers. Detailed painterly illustration, copper and teal palette, soft workshop light, plain sepia backdrop, vertical frame.
Run this briefIllustrationIllustration made from this prompt with an image model, not a result from the UniqAI app.
Full-figure fantasy character art of a wandering druid: a lean young man with light brown skin, moss-green knotwork tattoos on his forearms, a cloak of layered leaves, a staff topped with a living sapling and a fox cub at his feet. Painterly tabletop character-portrait style, dappled woodland light, earthy palette, plain parchment backdrop, vertical frame.
Run this briefIllustrationIllustration made from this prompt with an image model, not a result from the UniqAI app.
Full-figure character design of a near-future city courier: an East Asian woman in her twenties with a shaved undercut and a blue-dyed fringe, reflective yellow jacket, armoured knee pads, a messenger bag with glowing straps, mid-stride. Clean cel-shaded illustration, magenta rim light, plain dark grey backdrop, vertical frame.
Run this briefHow do I write a character description that renders well?
Treat the prompt as a character sheet, in this order:
- Who: species or age, build, skin, hair and one memorable feature (a scar, a prosthetic arm, round glasses).
- Outfit and gear: three or four concrete items, from head to boots. "Scaled leather armour, a longbow on her back" beats "cool armour".
- Palette: two or three named colours keep the design coherent.
- Pose and expression: "standing relaxed, weight on one leg" or "mid-stride, looking over the shoulder".
- Style and backdrop: "painterly concept art", "cel-shaded" or "storybook gouache", then "plain grey backdrop, full figure, vertical frame".
One character per image: two people in one prompt tend to swap features. Keep the description under about eighty words and change one line at a time between runs. The same recipe applies to any subject on the text to image page.
Will the same character look the same in every image?
Not exactly, and it is better to know that before you plan a comic. Each run composes the character again from the words, so the face, the cut of the cloak or the number of buckles can shift between images. There is no character-reference or saved-identity feature in the generator today. What helps: keep the exact wording and add only a new pose or scene line, fix the distinctive features in words ("thin scar through the left eyebrow, close-cropped hair"), generate several candidates and keep the closest ones, and use one style phrase for the whole set. For a character based on your own face, starting every run from the same photo on Nano Banana keeps the likeness far steadier.
Can I turn my photo into a character?
Yes. Upload a clear photo of yourself in the web app, pick Nano Banana and describe the character around your face: "turn me into a starship pilot in an orange pressure suit, painterly concept art, keep my face exactly as in the photo". It costs 12 gems per run and the likeness is kept, though fine detail can drift, so compare with the original. The AI art generator has ready prompts for oil, anime and comic looks on a photo, and AI avatars covers stylised profile pictures.
Is this character generator free, and what does a run cost?
Every image costs gems, and the price is shown next to the model before you run it: 3 gems for a Flux Schnell draft, 18 for a FLUX 1.1 Pro final on a paid plan, 12 for a Nano Banana run on your photo. If a generation fails, its gems are returned. A full first pass on a character, six drafts and one final, comes to about 36 gems. Plans and gem packs are listed on the pricing page, together with what a new account starts with.
About the examples
The characters on this page are illustrations made with an image model from the prompts shown under each card, so you can see what a given brief produces. They are not UniqAI outputs and all of them are original designs. Examples generated in the app will be added.
What it can and cannot do
It makes original characters from your words: heroes for a tabletop campaign, a cast for a novel, NPCs and concept sheets for a game. It will not reproduce a named character from a film, game or comic, and it is not built to draw real people from a description; describe your own design instead. Small text on badges, exact weapon mechanics and hands holding fine objects are the usual weak spots, and a long prompt with conflicting instructions renders less predictably than a short clear one. A finished character image can be animated into a short clip on the image to video page for a trailer or a social post.
Questions about AI Character Generator
Is there a free AI character generator here?
Each image costs gems and the price is shown before you run: a Flux Schnell draft is 3 gems, a FLUX 1.1 Pro final 18 on a paid plan, a Nano Banana run on your own photo 12. A failed generation returns its gems. The pricing page lists plans, gem packs and what a new account starts with.
Can I make D&D character art for my campaign?
Yes. Describe your own character, from ancestry and class to gear and palette, add "full figure, plain backdrop, painterly tabletop style" and run it. Describe the design rather than naming a published character or artwork, so the result is yours to use at the table.
Can I keep the same character across several images?
Only approximately. The generator has no character-reference feature, so each run redraws the design from the words and details can shift. Reuse the exact prompt, change only the pose or scene line, generate several candidates and keep the closest. For a character based on your face, start each run from the same photo on Nano Banana.
Can I design characters for a book or a comic?
Yes, for a cast page, cover concepts or reference sheets. Write one prompt per character, keep one style phrase for the whole cast and save the prompts with the images you pick. For sequential panels where the character must stay identical, expect to generate extra candidates and select by hand.
Can I animate the character for a video?
Yes. Save the character image, open the image to video page and upload it as the starting frame; Kling 3.0 and the other video models turn it into a short clip. Video models need a paid plan and cost more gems than an image, with the price shown before you run.
Coming soon to the App Store for iPhone
Chat, photo and video generation with several AI models in one app.