> For the complete documentation index, see [llms.txt](https://wiki.secrets.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://wiki.secrets.ai/create/generation-studio.md).

# Generation Studio

Choose your characters, build a scene, make images and videos, and organize your results in the Generation Studio.

Open **Generate** to create images and videos in one place. Choose **Image** or **Video** above the form, then use the gallery to review, reuse, edit, or animate your results.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fgit-blob-22ee5ee86ea7e1b309686c1affd488913c09e4b6%2Fstudio.webp?alt=media" alt="Generation Studio with Character, Action, Outfit, and Scene cards, an image prompt, size and quantity controls, and an All gallery containing photos and a video."><figcaption><p>Build your scene on the left and review your creations on the right.</p></figcaption></figure>

On desktop, the form sits beside your gallery and the Generate button stays within reach at the bottom of the form. On a phone, select **Gallery** beside Image / Video to open the gallery in a full-screen sheet. Close the sheet to return to your form.

## Make your first image

1. Select **Image**, then open **Character** and choose a companion.
2. Optionally choose an **Action**, **Outfit**, and **Scene**. You can also leave these empty and describe everything yourself.
3. Write your **Prompt**: explain what the character is doing, where they are, and the clothing, lighting, or mood you want.
4. Choose the **Aspect Ratio**, **Resolution**, and **Quantity** where available.
5. Read the total Moments shown on **Generate**, then select it to submit the request.
6. Find the result in the gallery. Open it to inspect the image and decide what to do next.

For example:

> Sitting by a cafe window with a warm drink, cream cardigan, soft morning light. Relaxed expression, waist-up portrait.

The companion's appearance reference supplies their likeness. Your prompt supplies the scene. For mode differences, reference-photo costs, and troubleshooting, see [Image Generation](/create/image-generation.md).

## Character, Action, Outfit, and Scene

| Card          | What goes here                                                                                               | Example         |
| ------------- | ------------------------------------------------------------------------------------------------------------ | --------------- |
| **Character** | Who appears in the image. Choose one companion, add more for a group, or choose a saved persona on Ultimate. | Lindsey         |
| **Action**    | An available pose or activity preset.                                                                        | A seated pose   |
| **Outfit**    | An available clothing preset.                                                                                | A casual outfit |
| **Scene**     | An available setting preset.                                                                                 | A cafe          |

Choose the character first so the other cards can show the available presets. Open a card to browse its choices; change or clear a selection when you want to try something else. Presets are optional and can be combined with a written prompt. Avoid asking for an outfit or setting that contradicts your selected preset.

Preset choices depend on the companion. Female realistic and anime companions have presets, and male companion presets are available too. Presets are not offered for trans companions or solo persona photos; describe those scenes in the prompt instead. The **Video** tab shows only the Character card: describe video action in the video prompt.

If a companion supports both **Premium** and **Advanced**, a mode selector appears. V2 companions use Advanced automatically. You do not need to find or enable a missing mode selector for those companions.

## Prompt tools

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fgit-blob-03aeb016b8a358f19091950bc4797359808ba32a%2Fimage-form.webp?alt=media" alt="Image prompt with reference-photo button, random-prompt dice, Tips, Enhance prompt switch, aspect ratio, resolution, quantity, and Generate total." width="480"><figcaption><p>Start with your own description or a prompt idea, then check the settings and price.</p></figcaption></figure>

* **Random prompt (dice):** fills in an editable idea. With one companion, existing Action, Outfit, and Scene choices help guide the idea. Group ideas use the selected people. This fills the form; it does not start a generation or spend Moments.
* **Tips:** opens guidance for the current kind of generation.
* **Enhance prompt:** helps expand a short scene description. Premium and Ultimate subscribers can switch it off below the prompt for closer control over their wording. There is no separate enhancement charge.
* **Reference-photo button:** opens the gallery picker. In Advanced mode, use reference photos to guide the look, clothing, pose, or setting. On desktop, you can also drag a gallery image onto the prompt.
* **Character counter:** shows how much room remains. Companion solo prompts allow 1,000 characters in Premium or 2,000 in Advanced; group prompts allow 5,000.

### Choose the size and quantity

**Aspect Ratio** controls the shape: use a portrait shape for a single person or a landscape shape for a wider scene. **Resolution** controls the image size. Advanced images offer 1K (Fast), 1.5K, and 2K; 4K is shown as coming soon. 1K and 1.5K have the same base price, while 2K costs more.

Use the **− / + Quantity** control for up to eight companion solo images. Group photos and generations from picked gallery references make one image at a time. On Ultimate, [persona solo photos](/create/persona-photos.md) also support 1–8 images using Quantity. Always read the total on Generate after changing resolution, quantity, people, or references.

While an Advanced request is running, you can prepare a different prompt or settings for another request. An identical request may stay disabled while it is already in progress; use Quantity when you want several versions of the same solo image.

## Put multiple people in a scene

Open **Character** and select a second companion to switch to a group photo. Add up to **three companions**. Selected companions stay at the top of the picker; select one again to remove them. Return to one selected companion for a solo photo.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fgit-blob-b605f08aa6a89d6d516ca44528cf4305cfc9059c%2Fgroup.webp?alt=media" alt="Two selected companions, Lindsey and Luca, and a group prompt using highlighted name mentions to give each person a different action." width="440"><figcaption><p>Use names to make each person's role in the scene clear.</p></figcaption></figure>

Type **@** in the group prompt and select a person from the menu, or select their name chip to insert it. For example:

> @Lindsey holds a folded map while @Luca points toward the harbor. A bright coastal afternoon, wide shot.

Give each person their own action, outfit, or position. A vague instruction such as “wearing a green coat” may be applied to the wrong person; “@Lindsey wears a green coat” is clearer.

**Ultimate:** use **You in the photo** to add a persona with a saved appearance photo. This is also available with a single companion, so you can make a photo of the two of you. The limit is **three people total**, including your persona. For a photo of only your persona, choose it under **Your personas** in Character instead. See [Persona Photos](/create/persona-photos.md).

## Turn a photo into a video

Open an image and select **Create Video**, or switch to **Video** and choose a source image. Describe the movement, camera motion, and sounds you want, then choose the model, resolution, and duration.

The model menu contains **Intimate**, **Cinematic**, **Everyday**, and **Basic**, with **Classic** under **More models**. **Fast** sits on the Cinematic row; **Turbo** sits on Intimate. See [Video Generation](/create/video-generation.md) for a comparison, frame controls, reference media, and prompting examples.

## Find and organize your results

| Gallery control             | Use it to…                                                                                  |
| --------------------------- | ------------------------------------------------------------------------------------------- |
| **All**                     | View photos and videos together. Videos have a play control.                                |
| **Images / Videos**         | Show only the selected media type.                                                          |
| **Recent / Favorites menu** | Change sorting, filter by tags, or open **Manage tags**.                                    |
| Selected character          | Narrow the gallery to that companion. With no character selected, browse your full gallery. |

Gallery tiles preserve the media's shape, so landscape images, portraits, and edited images can look different sizes. Pending generations show a placeholder until the result is ready.

On desktop, hover over a tile for quick actions. On a phone, tap the media to open its preview and actions.

<figure><img src="https://3286776735-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F9ACKOAU4nFWSXPeE2B4s%2Fuploads%2Fgit-blob-4ac92ecc4e330068b9f32b4b75a6dec4be491081%2Fimage-details.webp?alt=media" alt="Image Preview showing a cafe portrait, its prompt, size, creation date, tags, and Create Video, Reuse, Edit Image, Download, favorite, report, and delete actions."><figcaption><p>The preview brings the result, prompt, details, and actions together.</p></figcaption></figure>

* **Reuse:** restores the image's prompt and available creation settings so you can revise and generate again. It does not generate immediately. For newer edited images, Reuse reopens the edit with its original source and instructions, when that source is still available. Older edits may only restore their text to the main form.
* **Edit Image:** make a specific change to the existing photo. Describe the change and what should stay the same; edits cost 50 Moments. See [Image Editing](/create/image-editing.md).
* **Create Video:** use the selected image as the starting point for a video.
* **Favorite and tags:** mark useful results and organize them for later. Use the gallery menu to find them again.
* **Download:** save the result to your device.
* **Publish:** makes an eligible image public. Review it before publishing. A public image can have a share link; a private image does not.
* **Report or delete:** report a problem with a result, or remove a generation you no longer want. Read the confirmation before deleting.

Opening a video gives you playback controls and available actions such as downloading, favoriting, tagging, and reusing its setup.

## If something looks missing

**The Generate button is unavailable.** Check that you have a character or valid source, a prompt or supported preset selection, and enough Moments. Also check for an identical request already running.

**I cannot find a result.** Check All / Images / Videos, the selected companion, tag filters, and sorting. For persona photos, select the persona whose photos you want.

**I cannot find the group switch.** Select more than one companion in **Character**. For one companion plus your persona, use **You in the photo**.

**I cannot see a preset or model.** Available options depend on the selected companion and whether you are making an image, persona photo, or video. Use the guides for [images](/create/image-generation.md), [personas](/create/persona-photos.md), and [videos](/create/video-generation.md) for their specific options.