π Heads up β finish the previous lesson first. The course works in order.
Phase R Β· The Premium Studio Β· β± 20 min
The Premium Shoot
βΆ {{VIDEO: lesson r-3 β Owen & Cole}}Video drops in here. Everything below stands on its own.
When you're done
You'll have run her first premium photoshoot in the Create suite β preset pill + the six-part prompt formula, four takes for ~80 credits β and you'll have picked the keeper that anchors the rest of this section (it becomes the edit source in r-4, and the edited variant becomes the video start-frame after that).
Why this matters
The Content Factory from Phase 3 (3-1) keeps her feed alive daily with SFW content. This is the other engine: the premium content her Phase 6 ladder actually sells. Same discipline β recipes, batches, keepers β but a different studio with far more range, and a prompting culture worth stealing: on Rogue, the pros click a preset pill first and type second. The presets are professionally tuned per model; your prompt only adds the specifics the preset doesn't cover.
One mindset carry-over from 3-1 still rules: you are producing sets, not browsing possibilities. Decide the shot, run it, pick keepers, get out. The meter thanks you.
The steps
Select her, not a generic model. Create suite β Model Drawer β Character Models β My Characters β her card. Her panel opens showing her reference face, her image model (Softcore Image), and the aspect-ratio picker. This is the fidelity lock β every render in this session is her.
Lena selected β her face on the panel, Softcore Image as her image model. Every render from here is her.
Pick the scene from a preset pill. Above the prompt box: her model's preset menus (Clothes, Location, Selfies, Solo Play, Situationsβ¦). For the first premium set, go Selfies β Mirror Selfie (Night) β it reads amateur-real, which converts better than studio-perfect for a girl whose feed is built on "girl next door", and it's naturally suggestive without being explicit. You'll go further down the menus when her ladder calls for it (r-5).
The Selfies preset menu. Every entry is a pre-tuned professional prompt β click first, type second.
Add the formula prompt. The preset sets the shot; your prompt fills in the specifics using Rogue's six-part formula β subject, clothes, location, time, style, angle. The exact prompt from our live shoot:
she wears an oversized white tee slipping off one shoulder, hair up in a messy
bun, warm lamp light in her bedroom at night, amateur flash photography
Notice what's not in it: her face, her body, her identity β the character model carries all of that. You're only writing the scene. (And per r-1's rules: "woman" if you reference her at all, never "girl".)
Preset pill applied + formula prompt typed. The preset's preview shows what the pill alone would give you.
Check the button, then create. Generate showed 80 credits for this set β four takes. That's the read-the-button habit from r-1 in action: you knew the price before you clicked.
Four takes rendering. The filmstrip fills as they land.
Pick the keeper β same 10-second rule as Phase 3. Four takes came back; ours nailed the brief on take one: the tee, the bun, the lamp, the flash, and her face β cross-platform fidelity confirmed. Favorite the keeper (β₯) so it's easy to find in the library, and be honest about the misses.
The keeper, selected from the strip. Same girl as her Fanvue feed β that's the character model doing its job.The keeper, full resolution. This exact frame becomes the edit source in r-4.
Steal instead of guessing β the Inspire library. Before you invent shot ideas from nothing, open Inspire: a library of finished community renders you can remix β it loads the exact prompt, model, and settings behind any image, and you swap in your character. It's the single fastest way to expand her range without burning credits on experiments. (Same spirit as the Telegram metadata drops from r-1.)
Inspire: finished renders with their settings attached. Remix with her character instead of prompting blind.
AI prompts for this lesson
Copy these into your AI assistant when you want a hand. They know this lesson.
Want a premium scene menu in her voice? Same trick as 3-1:
My AI creator's persona: [paste sheet]. Suggest 5 premium photo scenes for her
suggestive-but-not-explicit tier, each as: one Rogue preset-pill idea (selfie /
situation / location flavor) + a one-line prompt in the formula "clothes,
location, time of day, style, angle". Amateur iPhone realism, no studio looks,
nothing explicit β this is the tier that teases the ladder.
Do it now
~20 minutes, ~80 credits.
You're done when
One premium keeper of her β unmistakably the same girl as her Fanvue profile β sits favorited in her Rogue library, and the set cost you double digits, not hundreds.
If you're stuck
The output doesn't look like her
Confirm you selected her under Character Models β My Characters β not a lookalike public model, and not a base model with her name typed in the prompt. Typing her name does nothing; the character model is the fidelity.
Which preset should I actually start with?
The one in the lesson. Mirror Selfie (Night) is deliberately chosen: amateur-real, suggestive, ladder-friendly. Your second set can come from the AI scene menu above.
My prompt got blocked
"Woman" not "girl"; no celebrity names; keep anatomical language plain. You weren't charged β reword, rerun. Playbook p41 for anything weirder.
Four takes, zero keepers
Rerun once with the same recipe (rendering variance is real), and check your prompt for contradictions with the preset (e.g., "beach at noon" under a night selfie pill). Still bad after two runs β change the preset, not the formula.
Should I shoot a whole premium week right now?
No. One set. The premium cadence gets planned in r-5 against her drop calendar β shooting inventory before the pipeline exists is how credits die.
Next up
The keeper is good. Next lesson makes it plural: one agentic edit turns the shot into a second sellable frame for 70 credits, and then the Motion suite makes her move.