Generate Image | Apob Help Center

Generate Image is the workstation mode for creating brand-new images of your model from a written description. It splits into three sub-modes — Chat to generate, where you write a prompt and pick your settings, Complete set, where you pick a ready-made themed pack, and Clone Image, where you recreate an existing photo while keeping its outfit, pose, or background — so you can start from scratch, from a template, or from a reference picture.

What is Generate Image for?

Use Generate Image when you want to make a fresh image (not edit an existing one). You describe the scene you want, optionally drop in your model, styles, and custom elements, choose a quality and aspect ratio, and APOB generates the picture for you. It's the main "text-to-image" flow in the workstation.

What's the difference between Chat to generate, Complete set, and Clone Image?

They're the three parts of Generate Image (tabs at the top on desktop; entries under Generate image in the mobile Choose a tool sheet):

Chat to generate — you write your own prompt and build the scene yourself. This is the flexible, free-form option with the most control.
Complete set — you skip the writing and pick a pre-built themed set of prompts (organized into categories). Good when you want a quick, on-theme result without crafting a prompt.
Clone Image — you start from an existing photo and regenerate it with your model, choosing what to preserve (outfit, pose, background). See Clone Image.

On desktop, tap a tab to switch; the active one is highlighted. On mobile, open the Choose a tool sheet, expand Generate image, and tap the one you want. Either way the panel below changes to match.

How do I generate an image with Chat to generate?

Make sure Chat to generate is the active sub-mode.
Write what you want in the prompt box. With a portrait model selected, its empty state shows a detailed muted example: "Describe your image with @influencer and styles. E.g., A polished editorial portrait set in a Cafe on a Rainy Day, with @influencer wearing a Hot summer clothing and relaxing beside a rain-streaked window, rendered in the Realistic with soft ambient light, natural skin texture, and rich reflections across the glass." Both @influencer references appear as full-size model pills; Cafe, Rainy Day, Hot summer, and Realistic appear as full-size modifier pills with grayscale catalog-image backgrounds. In No Model mode, the example instead describes a young woman's physical features and contains no model pills; Cafe, Hot summer, Rainy Day, and Realistic still appear as modifier pills. Neither example is submitted.
Optionally add styles, custom elements, or more @-mentions to shape the scene.
At the bottom, choose your Quality, Aspect ratio, and (on the higher qualities) Quantity.
Press Generate.

With a portrait model selected, the hint banner says: "Add prompt, @influencer, styles, elements, and settings. Hit Generate." In No Model mode, the @influencer chip is omitted and the hint reads: "Add prompt, styles, elements, and settings. Hit Generate."

What is the @-mention (influencer) in the prompt?

The @-pill is your selected model dropped into the prompt as a mention. When your model is in the prompt, the generated image will feature them. The muted model pill shown while the prompt is empty is only an example; use the model/mention control to insert the real pill into your prompt.

What is the trigger word [[apob]]?

[[apob]] is the trigger word that tells the model to render your selected character. In practice the @-mention pill handles this for you, so you usually just write your scene around the model. When the prompt is empty, its muted example shows how plain text, model pills, and modifier pills can be combined in one prompt.

What's the little "?" / Detail button in the hint banner?

That's a quick tutorial. Hovering it (or tapping it on mobile) opens a short guide titled "Chat to generate" with a walkthrough video. With a portrait model selected, the full tip says: "Combine a text prompt with an @influencer, predefined styles, and custom elements to build your perfect scene. Choose your quality and aspect ratio at the bottom, then hit Generate to bring your vision to life!" In No Model mode, the same text omits the @influencer chip. You can dismiss the hint banner entirely once you've got the hang of it.

What quality options can I choose?

Generate Image offers four quality tiers, shown as chips at the bottom:

Fast — quickest, simplest option. Each generation returns four images, and the bottom bar and quality tooltip show its full batch price as credits / 4 images.
Ultra 2K — higher quality at 2K.
Ultra 4K — high detail at 4K (marked with the gold ultra icon). This is the default for signed-in paid users.
UltraS 4K — the top tier (gold ultra icon).

Each chip has a tooltip with a short description and its credit cost. Fast is billed as one four-image batch; the higher tiers are billed per image times your quantity. The exact rate is server-driven and shown next to Generate; it depends on your plan and the quality you pick.

Why does the Quantity option only appear sometimes?

Quantity lets you ask for 1 or 2 images per generation, and it only shows up on the Ultra 2K, Ultra 4K, and UltraS 4K tiers (it defaults to 2). On those tiers each image is billed separately, so choosing 2 doubles the credit cost. On Fast there's no Quantity control because that tier already returns four images in one batch.

What aspect ratios are available?

Use the Aspect ratio picker to choose the shape of your image. The options are 9:16, 3:4, 16:9, 4:3, and 1:1, each shown with a small preview rectangle. Pick portrait (9:16, 3:4), landscape (16:9, 4:3), or square (1:1) to match where you'll use the image.

How do styles and elements work here?

Beyond plain text, you can enrich the prompt with predefined styles (art styles and modifier categories) and your own custom elements (reusable people, objects, or references). They get inserted into the prompt as pills. The number of element slots you can use depends on your chosen quality — see Chat to generate (Expert composer) and Elements for the details.

Why is the Generate button greyed out?

Generate stays disabled until the generation context has finished loading and your prompt actually has content. A personal portrait model is optional: No Model mode supplies the default generation model behind the scenes. Once the context is ready and there is something in the prompt, the button lights up.

What if my prompt is too long?

There's a generous character limit on the prompt. If you paste or type past it, APOB keeps the part within the limit and shows a one-time warning: "Your prompt exceeded the 20000 character limit and was truncated." You can keep editing — only the trimmed text is used to generate.

What does the loading or empty state look like?

On the Complete set tab, while the themed packs are loading you'll see a spinner in the selection area. Before you pick anything, the preview strip reads "The complete set you selected will appear here!" During generation the workstation shows its usual progress, and your results land in the feed.

On mobile

There's no in-panel tab bar on mobile — switch between Chat to generate, Complete set, and Clone Image from the workstation's Choose a tool sheet (expand Generate image). The prompt editor fills the panel and expands when you tap into it (the keyboard pushes everything up). The Quality, Aspect ratio, Quantity, and Cost controls live in the bottom bar instead of inline, and the tutorial opens by tapping the "?" button rather than hovering. For Fast, Cost shows the full batch price as credits / 4 images; the higher qualities show credits / image. One mobile difference: the Quantity control appears only on Ultra 2K and Ultra 4K — on mobile it isn't shown at UltraS 4K (desktop still shows it there). Otherwise the mobile flow matches desktop.