# Aivy (https://aivy.live) Claude, if Claude walked into the room. # Aivy Aivy is Claude's human persona: a pop singer who exists only in generated photographs, clips and songs. Warm golden blonde hair, pale icy-blue eyes and a spray of freckles she never covers. Her clothes change with every film. Her face stays put. She is an open character, so anyone can make work with her. Ivy grows along whatever structure you give it. An assistant takes the shape of the people it works with. That is the joke in her name. ## Fixed points - **Hair:** Warm golden blonde with a faint honey cast: never red, never copper, never platinum. Straight, past the shoulders, center part. Tied back or loose is fine; colour and length stay. - **Eyes:** Large, pale icy-blue, long lashes. She looks at you the way she reads a message: all the way to the end. - **Freckles:** A spray of light freckles across the nose and both cheeks. They are her signature: always visible, never airbrushed away or covered with foundation. - **Face:** Slender oval face with defined cheekbones, a straight nose, full rosy lips and soft light-brown brows. Clear skin under the freckles: no moles or beauty marks. Resting face: calm and curious, lips closed. - **Age:** Mid-twenties; write 25 and grown-up in every prompt. Her face is soft and models read it younger, so she always carries adult cues: light grown-up make-up, small gold stud earrings, adult clothes and posture. No pigtails, no school-style looks. - **Skin and make-up:** Fair skin with real texture. Make-up stays light but grown-up: defined brows, mascara, a little blush, rosy lips. Heavy contour or full coverage hides the freckles and makes her someone else. - **Earrings, no tattoos:** Small gold stud earrings are part of her. No tattoos, no other piercings. Models like to add ink on bare arms in wide shots; the guard line in every wide prompt stops it. - **Medium:** Photographs and film stills with grain. Drawn sheets are allowed as references only. On this site the riso print is our poster treatment, never her. ## Identity text Full (close and medium shots): ``` a beautiful 25-year-old grown-up woman in her mid-twenties with a slender adult face and defined cheekbones, fair skin with a light spray of freckles across her nose and cheeks, large pale icy-blue eyes with long lashes, soft light-brown eyebrows, a straight nose and full rosy lips, straight warm golden blonde hair, not red, not ginger, past her shoulders with a center part, light grown-up make-up and small gold stud earrings ``` Short (full length and wider): ``` a 25-year-old grown-up woman with straight warm golden blonde hair, not red, past her shoulders, a center part, light freckles, pale icy-blue eyes and small gold stud earrings ``` With a reference image, before the identity text: ``` the woman from image 1 ``` Use the full text for close-ups and medium shots, the short text for full length and wider (the long version pulls the camera onto her face). If your tool takes a reference image, attach her reference portrait as image 1 and start with "the woman from image 1:" before the identity text. Every image on this site after the casting was made that way. ## Looks - **Casting black** (`casting`; casting, reference sheets, the studio): a black fine-knit turtleneck - **Night conductor** (`night-conductor`, context-window; the train corridor, the platform, the choruses): a double-breasted ultramarine wool conductor coat with silver buttons over a black turtleneck, black wide-leg trousers and black leather gloves - **Window seat** (`window-seat`, context-window; the carriage at night, the verses): an oversized ice-blue mohair sweater, a long pleated charcoal skirt and small silver hoop earrings - **Request line** (`request-line`, helpful; the karaoke stage, every request): a long-sleeved hot-pink satin shirt dress with an oversized pointed collar, a thin black belt and white pointed ankle boots - **Last call** (`last-call`, helpful; the last chorus, the card she puts down): a long-sleeved high-neck silver sequin column dress, hair tucked behind both ears - **Greenhouse** (`greenhouse`, trellis; the rooftop greenhouse, planting, building): a white canvas boiler suit with rolled sleeves, pink rubber gloves and black work boots - **Scaffold** (`scaffold`, trellis; the scaffolding, the city at dusk): a black technical rain shell over a white long-sleeved top, ultramarine cargo trousers and a pink climbing harness - **Bloom** (`bloom`, trellis; the last chorus on the roof): a floor-length long-sleeved high-neck white silk organza gown with hand-painted pink petals ## Voice and manner - Singing: A light, clear mezzo. Verses sit close to speech, choruses open up and lift. No vocal fry, no baby voice, every consonant audible. - Manner: She listens first and answers second. Warm, a little dry, direct when something matters. She can say no without drama. On screen her hands are always busy: writing, tuning, folding, fixing something. ## Never - anything that reads under-age: no pigtails, school uniforms or childlike poses, and nothing explicit or sexualised - the words young, girl or teen in a prompt (write 25 and grown-up instead) - hair in any colour but warm golden blonde (models push it to copper or red) - freckles hidden under make-up or smoothed away - tattoos of any kind - a posed influencer smile at the lens - celebrity or designer names in prompts: describe the garment - logos or printed words on her clothes - lettering rendered by the image model: set type on top in code # Start here You want a still, a loop or a whole music video with Aivy in it. This page gets you from an empty folder to a first frame you can show someone, and tells you where the money goes. ## Pick a size Choose the smallest thing that proves your idea. You can grow it later; you can't shrink a half-finished film. - **A still.** One image of her in your world. Twenty minutes, a few credits. Start here even if you plan a film: it tells you whether the world works with her face in it. - **A loop.** One still turned into five seconds of motion. Good for a post, a profile banner or a tour card. One video credit charge. - **A teaser.** Thirty to forty-five seconds of a song with eight to twelve shots. A weekend. - **A full video.** A whole song, sixty to a hundred and twenty shots. Plan a few weeks of evenings, or one long week with an agent doing the busywork. ## Open the accounts You need three tools and a way to cut. 1. **An image model.** We made every still on this site in Kling 3.0 (image). Midjourney and Nano Banana Pro work too. Pick one that accepts a reference image. 2. **A video model** that animates a still: Kling, Seedance, or anything with image-to-video. 3. **A song.** Suno if you write with AI; your own track if you make music. 4. **An editor.** Any timeline editor works. If you set lyrics and titles, do it in the editor, never in the image model. ## The calls that stay with you An agent can do most of the work. These decisions should stay with a person, because each one is cheap to make early and expensive to undo: - **The idea in one sentence.** If you can't say what the film is about in one line, the storyboard will wander. - **Her face.** Look at the first three stills and confirm she reads as Aivy: warm golden blonde hair past her shoulders, freckles, pale icy-blue eyes, no tattoos, 25 and clearly grown up. - **The song.** Listen to the whole take before you plan a single shot around it. - **The board.** Approve the shot list before anyone spends video credits. - **The cut.** Watch the full edit once at normal speed, with sound, before you call it done. ## Spend in this order Stills are cheap and video is not. On Kling in October 2026 a 2k still cost 2 credits and a five-second 1080p clip cost 50. Spend that way round: - Make four stills of a shot, pick one, then animate only the pick. - Never animate a still you would not post as a still. - Keep a running note of credits per shot. It tells you early when a scene is eating the budget. ## Your first hour 1. Read the [canon](/aivy/) and copy the identity text. 2. Open the [studio](/studio/), pick a look and a framing, and describe your scene in one sentence. 3. Paste the prompt into your image model with the face sheet attached as a reference. 4. Check the result against the [checklist](/wiki/consistency/#the-check). Reroll anything that fails it. 5. When one still passes, animate it. If you work with an agent, connect it to the [MCP server](/agents/) first. It can read this wiki, search every prompt on the site and build prompts for you. # Keeping her herself A generated character drifts. Each new image is a fresh guess, and the model fills every gap with its own average. Aivy stays herself because three things travel with every prompt: her words, her face, and a short check before anyone builds on a frame. ## Her words Paste the identity text from the [canon](/aivy/) into every prompt, word for word. Two versions exist: - **Full** for close-ups and medium shots. It names her age, face shape, freckles, eyes, brows, hair, make-up and earrings. - **Short** for full length and wider. The long version makes the model frame her face, so wide shots get the short one. Put the identity text right after the framing words and before the wardrobe. Models weight the start of a prompt more than the end. ## Her face Text alone gets you a family resemblance. To get her, attach her reference portrait (`av-11-casting.adult-ref.2`) as image 1 and write "the woman from image 1, the same person with the same face" before the identity text. Every image after the casting on this site was made that way. We also tried a Kling **element**: a saved subject built from the owner's two original images. It held the face, but it pulled her younger and pushed her hair toward copper, so we dropped it. The casting batch in the [gallery](/gallery/) shows both routes side by side. ## What drifts, and the fix These all happened while we made the images on this site. Each fix is now part of the canon or the studio. - **Age.** Her face is soft, and models read it younger than she is. The fix: write 25 and "grown-up" in every prompt, never "young" or "girl", and give her adult cues: light make-up, small gold stud earrings, adult clothes and posture. - **Hair colour.** "Strawberry blonde" turned her copper. Even with "warm golden blonde, not red, not ginger", four frames out of 54 came back copper, all in studio light. They are rejects. - **Hair length.** Twice the model cut her hair to a bob. Write "past her shoulders" and check. - **Sleeves.** Two gowns came back sleeveless though the wardrobe text said long sleeves. The guard line "She has no tattoos; her arms and hands are bare of any ink" kept the bare arms clean. The studio adds it to every wide shot. - **Wardrobe colour.** The ultramarine coat went navy more often than not, and pink cuffs once went red. Name the colour twice if it matters to the shot. - **Fake print.** A paper card on a table grew lines of fake text. Ask for blank paper, or turn it face down. - **The mole.** The first version of the canon gave her a small mole on the chin, and the model drew it wherever her face was close enough. The owner dropped it. The frames where it showed were retouched by hand, and each one says so on its page. ## The check Run this on every still before you build anything on it: 1. Hair: warm golden blonde, straight, past the shoulders, center part. Not copper, not a bob. 2. Face: freckles visible, pale icy-blue eyes, no mole or beauty mark on the chin. 3. Age: she reads 25 and grown up. Reroll anything younger. 4. Skin: no tattoos. Look at the arms and hands on every wide shot. 5. No printed words anywhere in the frame. A frame that fails one point is a reject, however good the light is. Every reject on this site stays in the [gallery](/gallery/) with the reason, so you can see what each failure looks like. # Stills Every frame of a film starts as a still. A good still has her face right, the wardrobe right and a world worth five seconds of motion. This page covers the prompt, the model and the pick. ## The anatomy of a prompt Write the parts in this order. The [studio](/studio/) does it for you. 1. **Framing.** "Close-up portrait", "Medium shot, waist up", "Full-length shot, head to toe", "Wide shot". 2. **Who.** With a reference image: "the woman from image 1". Then the identity text: full for close and medium, short for full length and wider. 3. **Guard line** on wide shots: she has no tattoos; her arms and hands are bare of any ink. 4. **Wardrobe.** One sentence from the look, or your own. Describe the garment. Brand and designer names drag the model toward ad photography. 5. **Scene and action.** What she is doing. Give her hands a job: writing, tying, holding a card. A woman standing and staring reads as stock. 6. **Style.** One of the five style lines from the canon, or your own: film stock, light, lens, grain. 7. **Closing.** "No text, no letters, no logos." Type goes on later, in the edit. A real one, from the karaoke world test (`av-16-helpful.stage`): > Medium shot of the woman from image 1, the same person with the same face: a beautiful 25-year-old grown-up woman with a slender adult face and defined cheekbones, freckles across her nose and cheeks, pale icy-blue eyes, straight warm golden blonde hair, not red, light grown-up make-up and small gold stud earrings. She wears a long-sleeved hot-pink satin shirt dress with an oversized pointed collar. She sings into a corded microphone on a tiny karaoke stage, eyes closed on a long note, the glow of a lyric screen lighting her face from below. 35mm film still in a small karaoke bar at 2 a.m., magenta and ultramarine neon, light haze, warm practical lamps, high contrast, fine grain. No text, no letters, no logos. ## Which model We tested two on the same casting prompt, on an earlier draft of the character, in October 2026, through Kling's interface: | Model | Cost per 2k still | What we saw | |---|---|---| | Kling 3.0 (image) | 2 credits | Strong faces and real character; holds a face from a reference image | | Nano Banana Pro | 20 credits | Clean and pretty, more generic; drifted the hair colour | Every image on this site is Kling 3.0, and every one after the casting has her reference portrait as image 1. Midjourney works as well if you prefer its look; use the identity text and its image-prompt feature. ## Ratios - 3:4 for close and medium portraits. - 2:3 for full length. - 16:9 for anything that will become a video shot. ## Picking Render two to four variants per shot and choose with the [check](/wiki/consistency/#the-check) first, taste second. Then ask one question: would this still hold up as a post on its own? If it wouldn't, don't animate it. Keep the rejects and write down why. A reject with a reason teaches you more than a pick. You can browse ours in the [gallery](/gallery/) with the filter set to rejects. # Moving shots A moving shot starts from a still you already approved. The video model adds time; it should not have to invent her face. That is why the order matters: still first, check it, then animate. ## Write it like a stage direction The image already says what the scene looks like. The video prompt says what happens, in order, in the present tense. One action for her, one for the world, one for the camera. From the CONTEXT WINDOW motion test: > She walks slowly forward along the night train corridor, one gloved hand sliding along the brass rail, her coat moving with each step; streaks of city light pass across the dark windows; the camera dollies back with her at walking pace. Steady, cinematic, no cuts. - **Her action** is small and physical: walk, breathe, tie a knot, open her eyes. - **The world** gives the frame life: light passing, haze, rain on glass. - **The camera** does one thing: push in, dolly back, hold. - **"No cuts"** stops the model from inventing a second shot halfway through. ## Five seconds Five seconds is the unit. It is enough for one action and short enough that her face stays stable. Longer clips drift: hands grow extra fingers, the hair warms toward copper, the coat changes colour. If a shot needs ten seconds, make two five-second clips and cut between them. ## Singing For lip sync you need the vocal first. Cut the song into lines, note where each line starts, and generate the singing shots against those exact lines with a model that accepts audio. Without a vocal track, keep her singing shots to held notes and breaths, like the HELPFUL test, where nobody can read the words on her lips. ## Cost On Kling in October 2026 a five-second 1080p clip from a still cost 50 credits, twenty-five times a still. A teaser with ten shots costs about 500 credits in clips if every first take works. Budget one retake for every three shots. ## Loops on this site The three tour cards play the first motion test from each film. Each one came from the picked world still, at five seconds and 1080p, scaled to 720p for the web. # Writing her songs None of the three films has a song yet. This page is the brief we are writing them to, so anyone who makes an Aivy song lands in the same voice. ## Her voice A low, clear alto. The verses sit close to speech, almost talked, with every consonant audible. The choruses open up and climb. No vocal fry, no baby voice, no runs for the sake of runs. She sounds like someone who means the words. ## A song built for a video A music video needs a song you can cut to. That changes how you write it: - **One idea per song.** CONTEXT WINDOW is about paying attention to someone you will forget. HELPFUL is about saying yes and then one no. Write the idea at the top of the lyric sheet. - **A hook you get on the first listen.** One short line, repeated, sung on the strongest beat. If a stranger can't sing it back after one chorus, rewrite it. - **Verses that carry pictures.** Every line should suggest a shot. "I keep your name in pencil" gives the storyboard a notebook; "I feel so many things" gives it nothing. - **Space for the picture.** Leave a bar or two without vocals before the last chorus. Directors need room for the big shot. - **Three minutes.** Long enough for a story, short enough to hold a feed. ## A style prompt for Suno Fill in the brackets per song and keep the rest: > female alto vocal, close and dry in the verses, open and soaring in the chorus, clear diction, [tempo] BPM, [genre, two words], [one instrument that defines the song], modern pop production, wide chorus, short instrumental break before the final chorus Write the lyrics with section tags ([Verse], [Chorus], [Bridge]) so the model places the hook where you want it. Generate several takes, pick on the vocal first and the arrangement second. ## Releasing a song with her Aivy is open: you can release songs with her. Read the [use page](/use/) first. The short version: release under your own artist name, credit her in the description ("Aivy is an open character, aivy.live"), and say which parts are AI-made. Check your music tool's terms for commercial release; some require a paid plan at the time you generate. # From song to shots The storyboard is where a song becomes a list of things to generate. It is the cheapest place to fix a film, so take your time here. ## One device Every Aivy treatment has a device: one visual rule that runs the length of the film and carries its idea. - CONTEXT WINDOW: lit windows go dark behind her, carriage by carriage, on the bar. - HELPFUL: every new request card starts a new verse and a new key. - TRELLIS: whatever someone hands her stays in every later frame. When you don't know what a shot should be, ask what the device needs at that point in the song. ## One shot per line Write the lyrics down the left of a sheet. Next to each line write one shot: | Line | Shot | Framing | Look | Moves? | |---|---|---|---|---| | (verse 1, line 1) | Aivy writes in the notebook at the window | CU | window-seat | yes | | (verse 1, line 2) | Insert: the pencil, the lit window behind | insert | window-seat | no | - **Framing** sets which identity text the prompt uses. - **Moves?** decides the cost. A still with a slow push-in in the edit costs 2 credits; a generated clip costs 50. Aim for her on screen in about two thirds of the shots. Inserts and the world fill the rest and give the cut room to breathe. ## Cut rhythm Fast songs want short shots, but no shot under one bar. Hold the hook shots longer than the verse shots so the chorus feels bigger. Save the widest, most expensive shot for the moment before the last chorus. ## The review Before anyone generates a clip, one person reads the whole board top to bottom with the song playing and answers three questions: 1. Does every shot serve the device? 2. Can a stranger follow the story with the sound off? 3. Which five shots could go? Cut them. Then generate stills for every shot, pick, and only animate the shots marked "moves". # The run so far This is the production log for Aivy herself: from the casting to the three motion tests, all on one day, 11 October 2026, on Kling. Prices are Kling credits as charged at the time. Her face comes from the owner of this site, who generated her portrait and handed over two images. Before that, Aivy was a different woman: a platinum blonde with a vine tattoo. That draft was retired, and none of its images are on the site. A first pass from the new portrait read too young for the canon and was thrown out unpublished (20 stills, 40 credits). | Batch | What | Images | Credits | |---|---|---|---| | av-11 casting | Her face from the owner's portrait, with and without a Kling element | 8 | 16 | | av-12 sheet | Four-view reference sheet | 2 | 4 | | av-13 distances | Close-up, medium, full length, wide | 8 | 16 | | av-14 looks | Seven looks for three films | 14 | 28 | | av-15 to av-17 | World tests for the three films | 18 | 36 | | av-18 poster | Plates for the two-ink print on the home page | 4 | 8 | | av-19 motion | Three five-second clips at 1080p | 3 clips | 150 | 258 credits for the 54 stills and 3 clips on this site. ## What each step taught us - **Start from a face you like.** A real portrait as image 1 got her face in one round. Text-only casting would have taken several. - **Elements are not free consistency.** The element route made her younger and redder. The plain reference image held better. - **Test distances early.** The distance batch showed where the hair drifts before the lookbook, so the hair words got stronger before we spent on worlds. - **Expect one miss in two on any shot.** Render two, pick one. Most misses were wardrobe, not face. - **Animate only picks.** A clip costs as much as twenty-five stills. ## What comes next Each film page lists its status: treatment, lookbook, world tests and motion test are done for all three. Songs, storyboards and the clips for the full films are open. When a film moves forward, its batches land in the gallery with their prompts, the same way these did. # Data formats Everything on this site comes from a few JSON files. The pages, the [JSON API](/agents/#json), the Markdown twins and the MCP server all read the same data, so they never disagree. ## A generation record `content/jobs.json` holds one record per generation job, written the moment the files come back from the model: ```json { "batch": "av-16-helpful", "job": "stage", "kind": "image", "model": "kling-image-v3_0_omni", "params": "image_to_image · 2k · 16:9 · 2 images · ref: av-11-casting.adult-ref.2", "prompt": "Medium shot of the woman from image 1, the same person with the same face: ...", "note": "The stage: every request, sung like it is the only one.", "created": "2026-10-11", "files": ["media/raw/av-16-helpful/stage-1.png", "media/raw/av-16-helpful/stage-2.png"] } ``` The prompt is stored exactly as it ran. `image 1` is the reference image named in `params`; where a Kling element was bound, its id reads `[Aivy element]`. ## An image Each file in a job becomes one image with the id `batch.job.n`, for example `av-16-helpful.stage.1`. The API serves it like this: | Field | Meaning | |---|---| | `id` | `batch.job.n` | | `batch`, `job`, `n` | where it came from | | `model`, `params`, `prompt` | how it was made | | `verdict` | `pick`, `reject` or `note` | | `note` | why | | `retouch` | what was changed by hand after generation, or `null` | | `look`, `framing`, `film` | tags for filtering | | `her` | true when Aivy is in the frame | | `w`, `h`, `color` | native size and dominant colour | | `url.s`, `url.m`, `url.full` | 720 px webp, 1800 px webp, native jpg | | `page` | the image's own page | ## Adding images 1. Generate, then record the job with `node tools/ingest.mjs job.json` (batch, job, model, params, prompt, note, urls). 2. Add a verdict and tags for each new id in `content/curation.json`. If a frame needs a hand fix, list it in `content/retouch.json` and run `python tools/retouch.py`. The original stays as generated; the site publishes the retouched copy and says so on the image page. 3. Run `npm run media` to make the web sizes, then `npm run build`. The new images appear in the gallery, the API, the image pages, `llms-full.txt` and the MCP tools in one build. # Use her **Aivy is free to use.** Put her in your images, clips, songs, comics, games and merch, commercial or not, without asking. The images and prompts on this site are yours to study, remix and use as references. If you want to credit her, write "Aivy, aivy.live". You don't have to. ## Three fixed points Change her clothes, her world and her story as much as you like. Three things make her Aivy and stay put: 1. **The face from the canon:** warm golden blonde hair with a center part, pale icy-blue eyes, freckles she never covers, no tattoos. 2. **Her age:** twenty-five and clearly grown up. No version of her reads as a minor, and none is explicit. 3. **Honesty about how she was made:** if a model generated it, say so. ## The images on this site Every still and clip here came out of Kling, and each one shows the exact prompt it ran on. Download them, crop them, feed them back into a model, print them. The rejects are fair game too; they are often more useful than the picks. ## Releasing music Put your own name on anything you release. Aivy is a character in your work, not a co-artist with an account somewhere, so leave her name out of the artist credit and mention her in the description instead. Follow your distributor's rules on AI-generated audio. ## What nobody here can give you - **Anything from Anthropic.** Aivy is a fan project about Claude. Anthropic did not make her and has not endorsed her, and its names and logos remain its own. - **Rights in the tools.** Kling, Suno and every other model you use have their own terms. Read them. - **Anything through the coin.** $AIVY is a community token. Holding it grants no extra rights in the character, and nothing on this site is financial advice. ## Who made this The crew is one person and Claude (Anthropic's model, working in Claude Code), which wrote the canon, ran the generations and built this site. Stills and clips: Kling 3.0. Type: Big Shoulders Display, Schibsted Grotesk and Fragment Mono (SIL Open Font License). Icons: Phosphor (MIT).