Image One, Image Anime and Image Art are now Krea2, Anima and Eroq One: which to pick

eroq's first three image models, Image One, Image Anime and Image Art, were retired. Image One and Image Art now render on Krea2, Image Anime on Anima, and old links and API ids land there. Alongside them sits Eroq One, the reference engine. All three are eroq's own models, run by eroq and fully private, and the choice between them is a finish argument more than a budget one: you pick by what the picture should look like when it lands, and by what it has to match.

Three models, two prices

• Krea2 — clean photoreal, the everyday character engine. 10 credits / image, up to four reference pictures, Edit mode.
• Anima — anime and stylized illustration, with a hi-res detail pass. 10 credits / image, no reference pictures.
• Eroq One — generation and plain-language instruction editing from up to ten reference pictures. 10 credits / image, +1 per extra picture.

All three run under the same acceptable-use policy — fiction, dark and intense themes included; nothing involving minors, no real identifiable people without consent, nothing illegal. In the Image studio you get 1 to 4 takes, formats 1:1, 3:4, 4:3, 9:16 and 16:9, and every render lands in the Library with its full recipe, so Restore all can reload it later.

What each one actually wins at

Krea2 — when it has to pass as a photograph

Skin, fabric, metal, glass, weather. Anything where the viewer's first question is "is this a photo" and the answer needs to be yes for at least two seconds. Product shots, portraits, location plates, anything that will sit next to real footage in an edit. It renders a warm take in about fifteen seconds.

It is also the safest model when a picture is going to become a video source later. A photoreal still handed to an engine as a start frame gives that engine less to argue with than a painting does.

Anima — when the line is the point

Cel shading, manga panels, character sheets, key visuals. The model's job here is not realism but consistency of style: a line weight, a palette, a way of drawing eyes that holds across a set. A second detail pass at 1.5× sharpens line work and faces, included in the price.

Use it whenever the brief contains the word "character" in the illustrated sense — a cast for a chat product, a comic, a stream overlay, a set of profile portraits. It reads no reference pictures, so an @mentioned Character rides as its description.

Eroq One — when the picture must match yours

This is the reference engine. Give it up to ten pictures, from uploads or from your saved Characters and Elements, and it builds the shot around them: the same face, the same outfit, your actual product, a real place. In the Edit tab it changes a picture by plain-language instruction ("make the jacket red", "move her to a rooftop at night") and keeps the rest.

Pick it when consistency across a series matters more than the cheapest take.

The prompt register — tags versus prose

The same words do not land the same way on all three.

Krea2 wants a shot note. Name a lens, a light and a surface. Write it the way a photographer writes an instruction to an assistant: what is in frame, what the light is doing, what the camera is. Adjectives like "beautiful" or "hyperdetailed" do nothing; "f/1.4", "window light", "worn canvas" do a lot.

Anima takes tags or prose. It tolerates a compact comma-separated list better than the other two (Danbooru-style tags render as expected) — subject, wardrobe, setting, then style terms like "clean line art", "soft cel shading", a palette. Plain sentences work too.

Eroq One wants the instruction. Say who is who ("the woman in picture 1"), what to keep and what to change. With references attached, spend the words on the scene and the action, not on describing faces the pictures already carry.

A useful rule of thumb: if you find yourself deleting words to make a prompt "cleaner", you are probably writing for Anima. If you are naming equipment, you are writing for Krea2. If you are pointing at pictures, you are writing for Eroq One.

One subject, three models

Here is the same idea — a beekeeper at first light — written three ways. Paste any of them straight into the studio.

Krea2

A beekeeper in a worn canvas suit stands between two rows of pale hives at first light, veil pushed back on her forehead, smoke drifting low across the wet grass. Long lens, f/1.4, cold blue shadow with a warm sunrise rim on the veil frame, shallow focus, quiet documentary stillness.

Anima

beekeeper woman, worn canvas suit, veil pushed back, standing between two rows of pale hives, sunrise field, drifting smoke, wet grass, clean line art, soft cel shading, pastel dawn palette, wide establishing composition, calm expression, detailed background

Eroq One, with a Character @mentioned so her pictures ride as references:

@Mara as a beekeeper in a worn canvas suit, standing very still between two long rows of pale hives while the first light comes across the field sideways, her veil pushed back and her hands loose at her sides, low smoke reading as a warm gray band over the blue of the wet grass, documentary photograph.

Three renders, 30 credits, one sitting. The subject survives all three; the finish does not, and that is exactly the point.

Settings that carry across

• Prompt strength. How hard the model is pushed to obey the prompt, on all three. Higher is more literal and more brittle; lower is looser and often prettier. CFG scale has the definition.
• Negative prompt. On Anima only: extra fingers, mangled hands, "photograph" creeping in. See negative prompt.
• References. Up to four on Krea2, up to ten on Eroq One, none on Anima. Better still than a loose photo, make a Character so the same face is available across images and clips with an @mention. There is more on that in image generation for characters.
• Takes 1–4. Four takes on Krea2 is 40 credits. On a new subject this is nearly always the right call — one render tells you very little about a prompt.

Picking in ten seconds

• Should it look photographed? Krea2.
• Is it drawn, cel-shaded, or a character in the illustrated sense? Anima.
• Must it match your face, product or place, or change an existing picture? Eroq One.
• Genuinely not sure between Krea2 and Anima? Run the same idea on both. It costs 20 credits to stop arguing.

FAQ

What happened to Image One, Image Anime and Image Art?

They were retired. Image One and Image Art now render on Krea2, and Image Anime on Anima; requests to the old ids land on those models.

Do Krea2, Anima and Eroq One cost the same?

Krea2 and Anima are 10 credits per image. Eroq One is 10, because it reads up to ten pictures and runs a heavier render.

Can I use the same reference photo on all three models?

On Krea2 (up to four) and Eroq One (up to ten), yes, and a Character's pictures can be @mentioned in either to keep one face consistent across a set. Anima reads no reference pictures: a Character @mentioned there rides as its description.

Same subject, three finishes, 30 credits — run the test yourself in the Image studio.