How to Use Illustrations on My Spicy Vanilla
Turn any moment of a chapter into an image and keep your characters recognizable from one picture to the next. This guide shows you how to pick an art style and image model, illustrate a scene with Visualise, regenerate the chapter cover, edit an image and export the illustrated story.
Illustrations are AI-generated images of scenes from your chapters. My Spicy Vanilla reads the text, writes an image prompt that matches your characters, the setting and your chosen art style, and turns it into a picture you can place inside the chapter, use as a cover or download.
There are three places where illustrations appear. Every chapter gets a cover illustration automatically when it is generated. In Edit mode you can illustrate any moment of the chapter with the Visualise button. And the Add Image button in the editor lets you insert any image from your library. Generating images is a Premium feature; the style and model are chosen once in your settings, so that is the first step.
Step 1: choose your art style and image model
Go to Content Preferences → Images. The Illustrations section applies to every image you generate, in every story.
- Under Image Generation Model, pick a mode. Creative generates fresh artwork straight from each scene, for maximum variety. Character Consistency keeps your characters looking like themselves in every scene, using their reference portraits (see step 2).
- Under Model, choose an AI image model. Each card shows the estimated generation time, and models marked Premium need a Premium account. If you are unsure, keep the default.
- Under Image Style, click the style you like, for example Photorealism, Watercolor, Anime, Oil Painting or Dark Fantasy. Click See other styles to show them all. The previews change with the selected model, so you can compare before you decide.
- Click Update.
The Explicit Illustrations section has one switch, Enable Explicit Images, which allows nudity and explicit detail in your illustrations. It is a Premium setting and it applies to chapter covers. Scene illustrations made with Visualise follow the story instead: only stories with a spicy, erotic or kinky genre can get explicit scene illustrations.
Good to know: the Disable Story Images switch on the same page turns illustrations off completely. Chapters then generate faster and use text only, and the Visualise and Add Image buttons disappear from the editor until you turn images back on.
Step 2 (optional): give each character a portrait
A portrait is the reference image the AI uses to keep a character's face, hair and physical features the same in every illustration. Portraits are only used by Character Consistency models; a Creative model describes the character in words instead, so the look can change from one image to the next.
- Open Characters and edit a character. Scroll to the Portrait section.
- Click Change. The Change Image window opens with your image library; pick an image and click Assign, or use Generate new to create a portrait first.
- Repeat for the other main characters of the story. Up to 3 portraits are used in a single illustration.
The portrait is saved on the character, so it works in every story the character appears in. If your current model is not a Character Consistency model, a tip on the character page reminds you where to change it.
Step 3: illustrate a scene with Visualise
- Open the chapter from My Stories and click Edit at the top of the page.
- Select the passage you want to see as a picture, from one sentence to a few paragraphs. A toolbar appears under the selection; click Visualise. The button needs at least 3 selected words.
- While it says Generating prompt…, the AI reads the passage and writes an image description for it. You can change this text freely, up to 1,000 characters. Click Retry if you would like a different description.
- Under Characters in image, the characters the AI found in the scene are already ticked. Tick or untick them so only the people who should be in the picture are selected.
- Click Generate image. The panel says Visualising scene…; it usually takes under a minute, depending on the model.
- When the image appears, click Insert to place it right after the selected passage. Not happy with it? Regenerate image makes a new picture from the same description, Edit prompt takes you back to the text, and Discard closes the panel without inserting anything.
Examples of prompt changes that work well:
- Add the setting if it is missing: "a rooftop bar at night, city lights in the background".
- Describe the moment, not the whole scene: "Elena hands him the letter across the table, both of them smiling".
- Say what you want to see instead of what you don't: "clean-shaven" works, "no beard" often doesn't.
Plugins of the Image Generator type are applied when the description is written, so a rule like "outfits should match a 1920s aesthetic" carries over to every illustration. The Plugins tutorial explains how to set one up.
Place, resize and edit the illustration
Click an image inside the editor to open its toolbar:
- Full makes the image as wide as the text. Half places it on the left or the right half, with the text flowing next to it.
- Edit opens the Edit image window. Paint over the part you want to change with the Brush, or leave the picture blank to edit the whole image, then type an instruction such as "replace with a window overlooking the sea" and click Apply. Use Continue editing to make another change, or Try again for a different result.
- Delete image removes it from the chapter. The picture stays in your library, so you can insert it again later.
Image editing does not accept explicit instructions, and every edit is generated from the previous one, so the quality can drop after several edits in a row. If that happens, go back to the original and describe all the changes in one instruction.
You can also click Add Image in the editor toolbar to insert an image you generated earlier, or to generate a new one from the Generate new tab of the library.
Important: inserted images, like every other change in Edit mode, are only saved when you click Update at the bottom of the page. The AI Edit tutorial covers the rest of the editor.
Change the chapter cover
The cover is the large image at the top of the chapter page. Move the mouse over it (or tap it on a phone) to reveal two buttons: Regenerate image and Download image.
Regenerate image opens a window with three tabs:
- Generate new: describe what you want changed, for example Ashley should wear a red dress, tick Generate Erotic Image if the cover should be explicit, choose the characters under Which characters should be included in the image? and click Generate. The page refreshes when the new cover is ready.
- Edit current image: the same paint-over editing as in the editor, applied to the cover.
- Library: pick any image you generated before and click Apply.
Limits and common messages
- Premium only: generating and editing images needs a Premium account. Premium accounts can generate up to 100 images in any 24-hour period; when you reach the limit, a Get more images button lets you add a pack of 50 extra images for 24 hours.
- One at a time: only one image generation runs per account. If you start a second one, you are asked to wait for the first to finish.
- Moderation: every image description is checked before the image is made, and every finished image is checked again. A blocked description shows Your submission contains content that is not allowed with the reason; rephrase it and try again. If many of your images get flagged in one day, image generation is paused for 24 hours. The Image Moderation Policy explains what is allowed.
- Images may vary: the same description gives a different picture every time. If a detail is wrong, fix it with Edit image instead of regenerating, so you keep everything else.
- Regional restrictions: in some countries image generation is disabled by law. The settings page tells you if that applies to you.
If an image never arrives or the panel shows an error, close it, select the passage again and retry. For prompts the model keeps ignoring, see the troubleshooting guide.
Download, reuse and export
- Your library: every image you generate is kept on the Manage Images page. There you can browse by story, character or location, see which prompt, model and style were used, generate again, assign an image to a character or location, archive it or delete it.
- Download: use Download image on the chapter cover, or open any image from your library to save it in full size.
- Export: when you download a story as PDF, EPUB or HTML, the inserted illustrations are included. In the download window, Book cover picture lets you pick any image from your library as the book's cover with Choose cover.
Final tip
For your first illustration, keep the default model, pick one style you like and give your two main characters a portrait. Then open a chapter in Edit mode, select one short paragraph with a clear action and click Visualise. Read the generated description before you click Generate image: fixing one word there is faster than regenerating the picture. Once that works, switch to a Character Consistency model and compare how much more recognizable your characters become.
Love Reddit? So do we!
Join our new subreddit community to share tips, tricks, and experiences. Let's make the most out of My Spicy Vanilla together!