Dual-Persona Character Sheet Prompt Workflow for Gemini, Grok, and GPT Image
A comprehensive guide to generating a split-screen character concept board showcasing two distinct identities of the same person using Gemini, Grok, and GPT Ima
Shared by AI creator Seerat Fatima (@SeeratFatima112) on October 5, 2026, this practical workflow provides a 4-step framework and a comprehensive master prompt for generating split-screen character concept boards using modern multimodal image generation models like Google Gemini, xAI's Grok, and OpenAI's GPT Image 2.0.

Image source: @SeeratFatima112 / X
The technique visualizes two contrasting identities of the same character—such as a competitive football midfielder and a creative, stylish individual—arranged across an editorial-grade split layout complete with facial expression grids, statistical profile cards, fabric close-ups, color swatches, and handwritten quotes.
The 4-Step Character Sheet Generation Pipeline
The author's core pipeline relies directly on native reference image upload capabilities within contemporary generative image platforms, avoiding manual collage work or separate LoRA fine-tuning:
- Access the Model Interface: Open Google Gemini, xAI Grok, or OpenAI GPT Image 2.0.
- Upload Reference Photo: Attach a clear portrait photo to serve as the facial identity anchor across both personas.
- Input the Master Prompt: Paste the structured English prompt specifying split-screen layout, outfit styling, expression grids, and typography.
- Generate Images and Video: Produce the 8K editorial concept board, and optionally animate selected sub-panels into dynamic motion clips.
Structural Breakdown of the Dual-Persona Prompt
The effectiveness of this prompt stems from its clear spatial segregation and granular aesthetic direction across both persona zones.
1. Canvas Infrastructure and Common Aesthetic
The prompt establishes the canvas as a professional split-screen concept board titled "AARYA PATEL" set against a warm off-white grid background with elegant typography, cohesive color palettes, and editorial mood-board styling.
2. Top / Left Section: The Athlete Profile
Dedicated to athletic vigor and competitive focus:
- Main Shot: A 23-year-old Indian woman with mid-length wavy brown hair, wearing a black and gold team kit (Jersey #10) with matching shorts, socks, and cleats, seated outdoors next to a soccer ball and sports bag, flanked by the handwritten quote "Discipline Creates Freedom".
- Stats Panel: A structured data block listing role (Attacking Midfielder), personality traits, archetype (The Creator), height (178 cm / 5'10"), age (23), and nationality.
- Expressions Grid: A 2x2 grid of headshots labeled DETERMINED, HAPPY, FOCUSED, and CONFIDENT.
- Kit & Color Swatches: Inset detail shots of the gold crest, jersey number "10", sleeve detail, and fabric texture, with hex color swatches.
- Action Thumbnails & Vista: Miniature action vignettes (dribbling, shooting, celebrating) and a silhouette shot gazing over a sunset skyline inscribed with "Same Game, New Dreams".
3. Bottom / Right Section: The Creative Lifestyle Profile
Dedicated to thoughtful, relaxed elegance:
- Main Shot: The same woman seated in a monochromatic dusty blue ensemble (oversized knit sweater and wide-leg trousers), holding a notebook and pen in an ambient cafe or studio setting.
- Personality & Color Palette: Creative, independent, and ambitious attributes framed by warm beige, tan, soft blue, and dark chocolate swatches.
- Expressions Grid: A horizontal strip of soft expressions labeled HAPPY, THOUGHTFUL, PLAYFUL, SERIOUS, and CALM.
- Texture & Accessory Insets: Macro close-ups of soft knit yarn, minimal leather bag grain, and a delicate gold pendant necklace.
- Lifestyle Vignettes & Footer: Square candid photos (browsing books, holding coffee, strolling through a flower market) paired with cursive text "More To Come..." and a category marker "PEOPLE • STORIES • VISUALS".
Complete English Master Prompt
The full prompt text can be copied directly into Gemini, Grok, or GPT Image 2.0/2.5 without alteration, or adapted by substituting character names, demographic attributes, kit colors, and career roles:
A professional, modern split-screen character concept board titled "AARYA PATEL" featuring two distinct identities of the same young woman—a talented football player and a creative, stylish individual—displayed on a warm off-white grid canvas with elegant typography, color palettes, close-up details, and handwritten quotes.
Top / Left Section (The Athlete – Midfielder Profile):
* Main Shot: A full-body shot of a 23-year-old Indian woman with mid-length wavy brown hair, wearing a black and gold team kit (Jersey #10) with matching shorts, socks, and cleats. She is sitting outdoors on a concrete bench next to a soccer ball and sports bag. Handwritten text on the side reads "Discipline Creates Freedom".
* Stats Panel: Text block stating:
* Role: Attacking Midfielder
* Personality: Disciplined, Humble, Always Improving
* Archetype: The Creator
* Height: 178 cm (5'10")
* Age: 23
* Nationality: Indian
* Facial Expressions Grid: A 2x2 grid of headshots showing four distinct expressions labeled: DETERMINED, HAPPY, FOCUSED, and CONFIDENT.
* Kit & Color Details: Inset close-ups of the gold club crest, jersey number "10", sleeve detail, and fabric texture. Hex color swatches included.
* Action Shots: Small action thumbnails at the bottom showing her dribbling, shooting, and celebrating on the field.
* Bottom Vista: A silhouette/rear shot of her sitting with a football overlooking a city skyline at sunset, with handwritten text reading "Same Game, New Dreams".
Bottom / Right Section (The Creative – Lifestyle & Fashion Profile):
* Main Shot: A full-body and seated shot of the same woman in a stylish monochrome dusty blue outfit (oversized knit sweater with wide-leg pants / tailored blue suit). She holds a notebook and pen, looking thoughtful in a cozy studio or cafe setting with coffee and books nearby.
* Personality & Vibe:
* Traits: Creative, Independent, Kind, Focused, Ambitious, Emotionally Intelligent.
* Color Palette: Warm beige, tan, soft blue, and dark chocolate brown swatches.
* Expressions Grid: A row of headshots depicting soft expressions labeled: HAPPY, THOUGHTFUL, PLAYFUL, SERIOUS, and CALM.
* Texture & Accessory Details: Inset close-ups of soft blue knit fabric texture, minimal leather bag texture, and a simple gold pendant necklace.
* Lifestyle Shots: Small square photos showing her browsing a library, holding coffee, walking through a flower market, and standing in a trench coat.
* Footer: Handwritten cursive text reading "More To Come..." and a footer label "PEOPLE • STORIES • VISUALS".
Overall Visual Style: Hyper-realistic 8K photograph quality, warm and natural editorial lighting, clean graphic design layout, sophisticated font pairings, and a aesthetic mood board presentation.
Platform Considerations and Motion Extensions
- Facial Identity Consistency: Supplying an evenly illuminated, high-resolution portrait during Step 2 ensures the underlying vision encoder preserves facial geometry and features across both distinct wardrobe and lighting setups.
- Platform Capability Profiles:
- Google Gemini: Supports text-prompt image generation (
gemini-3.1-flash-image) and interleaved text-and-image outputs, alongside native video creation with reference photos (via Gemini Omni) and SynthID watermarking in the Gemini app. - xAI Grok (Imagine Image 2.0): Supports image editing with up to three reference images, providing built-in task templates for maintaining consistent character and asset styling across generations, accessible via web, mobile, and the
grok-imagine-image-2.0API. - OpenAI GPT Image (2.0 / 2.5): Offers
gpt-image-2.5-sunburstfor editing precision andgpt-image-2.5-flarefor fast everyday generation, supporting reference-based image editing and multi-turn iterative generation via the Responses API.
- Google Gemini: Supports text-prompt image generation (
- Video Motion Generation: The fourth step noted in the workflow incorporates video generation. Creators can take the generated concept board or crop individual sub-panels to serve as reference frames in compatible image-to-video tools—such as Gemini Omni, which accepts up to 5 reference photos and embeds SynthID verification—to animate action vignettes into dynamic video clips.
Original Source
- Seerat Fatima (@SeeratFatima112) on X: https://x.com/SeeratFatima112/status/2106928532048040014