“AA23 prompts” are AI image and video instructions built around one style: deep crimson-and-black neo-noir lighting, chiaroscuro shadow, and mythic hero energy. The name comes from the teaser hype around actor Allu Arjun’s 23rd film (directed by Lokesh Kanagaraj) — fans started feeding Google Gemini, Midjourney, and ChatGPT prompts that turn an ordinary selfie into a “mass hero” movie poster, and the trend spread across Instagram Reels and TikTok through 2026.
This guide collects the strongest, most visually distinct AA23 prompt variants we’ve tested — from still portraits to full running animations — with the exact copy-paste text, the tool each one works best in, and the settings that keep your face looking like you instead of a generic AI avatar.
What Is the AA23 Prompt Trend?
At its core, AA23 is a “style transfer” instruction set for AI image generators. It asks the model to keep your face and pose intact while replacing everything else — lighting, color grade, background, texture — with a gritty, comic-book-meets-cinema look. Three ingredients define it:
- Monochromatic crimson-red grading: the entire frame is pushed toward deep red and black, rather than a natural color palette.
- Chiaroscuro lighting: one strong light source (usually a glowing sun) creates hard shadows and dramatic silhouettes.
- Mythic or gangster symbolism: a roaring animal, a shattering stone floor, or a “mass entry” walk that signals power and inevitability.
Everything below builds on that same foundation, so once you understand one prompt you can mix and match elements from the others.
7 Best AA23 Prompt Variants (Copy & Paste)
1. Classic Neo-Noir Portrait
The foundation prompt. Best for a first attempt because it’s the most forgiving on any AI model.
Transform the uploaded image into a dark neo-noir cinematic illustration. Full-frame composition, entire subject and environment visible. Subject posed as: [standing tall, shoulders squared, calm dominant posture]. Monochromatic deep crimson red color grading across the entire image. High contrast lighting with heavy shadows and dramatic highlights. Cinematic chiaroscuro lighting with strong depth and mood. Gritty graphic novel movie-poster style with subtle halftone texture. Stylized digital painting with rough, expressive brush strokes. Dark psychological thriller atmosphere, intense, serious, powerful tone. Background: dense jungle with layered fog, a glowing yellow sun creating a silhouette effect. Stylized illustration, not photorealistic.
What makes it different: no animal, no motion, no gangster theme — just clean lighting and pose, which keeps facial likeness the most accurate of any variant here.
2. Mythic Stone Guardian
Adds a colossal stone-animal sculpture in the background for a “legend” feel rather than a simple portrait.
Transform the uploaded image into a dark neo-noir cinematic illustration. Subject posed as: [ADD POSE]. Monochromatic deep crimson and blood-red color grading, chiaroscuro lighting with strong silhouettes. Background: a vast jungle landscape, a massive glowing yellow sun low on the horizon. In the far background, a colossal black stone animal sculpture (lion, tiger, or bear) partially merged with cliffs, roaring, ancient and weathered, eyes catching light from the sun — visible but not overpowering the subject. Mood: primal dominance, destiny, inner fire. Stylized illustration, not photorealistic.
What makes it different: the stone-carved animal (instead of a living one) reads as ancient and mythic rather than aggressive — better suited to “legacy” or milestone posts than action content.
3. Tamil Gangster Mass-Entry Poster
The most popular variant for Instagram “character intro” reels, built around a confrontational stance rather than a calm pose.
Create a cinematic digital illustration of the person in the reference photo. Maintain the exact face identity, hairstyle, expression, skin tone, and pose with zero distortion. Transform the entire scene into a deep red high-contrast Tamil gangster movie poster style. Replace the background with red gradients, smoky atmosphere, and subtle textures. Add bold comic book shading, thick outlines, strong black shadows, and glossy highlights on skin and clothes. Gritty, powerful cinematic mood, dramatic lighting, high contrast. Ultra-detailed, 4K illustration.
What makes it different: bold comic-style outlines and glossy highlights replace the painterly brush-stroke texture of variant 1 — this one reads as a movie poster, not a digital painting.
4. God-Speed Running Animation (Video)
For Gemini/Veo-style video tools rather than static images — the subject is mid-motion, not posed.
Cinematic animation in a dark neo-noir mythic style. Full-body shot of the subject running at extreme speed across a long black stone landscape. Powerful, aggressive running motion with fast foot impact on rugged stone terrain. Deep crimson and blood-red color grading. High contrast lighting with intense shadows. Bright yellow speed lines and energy streaks forming behind the subject. Red and black motion blur trails. Ground cracking slightly under each step. Dust and debris exploding backward. Low-angle tracking shot. 4K, ultra-detailed, gritty graphic novel texture.
What makes it different: this is the only variant built for motion — yellow “speed line” energy trails and ground-cracking effects don’t appear in any still-portrait prompt.
5. Eye Close-Up Revenge Frame
An extreme close-up instead of a full-body shot — works well as a thumbnail or a mid-reel cutaway.
Transform the uploaded image into a dark neo-noir cinematic illustration. Extreme close-up framing focused on intense human eyes, sharp emotional gaze. Monochromatic deep blood-red color grading covering the entire image. High contrast lighting with heavy shadows and strong highlights. Graphic novel poster style with subtle halftone texture and film grain. Psychological thriller atmosphere, sense of rage, revenge, inner conflict. Cinematic composition, sharp focus on eyes, background fading into darkness. Stylized illustration, not photorealistic.
What makes it different: the framing itself (eyes only, background dissolved to black) is unique among these prompts — nothing else here crops that tight.
6. Hero Intro Reveal (Slow-Motion Walk)
Built specifically for the “reveal” moment in a reel, where the subject emerges from darkness rather than standing still.
Epic cinematic intro. Hero character emerging from darkness in a neo-noir environment. Slow-motion walk with powerful body language and intense aura. Deep red and black cinematic color grading with dramatic contrast. Heavy shadows and glowing highlights shaping the silhouette. Cinematic smoke, dust, and glowing embers surrounding the hero. Massive glowing sun or fiery light source behind the character. Mythic animal shadow forming in the background, symbolizing dominance. Dynamic low-angle camera movement with cinematic zoom and motion blur. Ultra-realistic cinematic lighting, dramatic atmosphere, heroic energy.
What makes it different: the subject is emerging from darkness (not silhouetted against a sun) and walking, not running or standing — a slower, more deliberate reveal than variant 4.
7. Duo / Couple Mythic Edit
The only two-person variant — swap the single-subject language for a pair and add a warmer light for anniversary or wedding-style reels.
Transform the uploaded image of two people into a dark neo-noir cinematic illustration. Maintain the exact face identity of both subjects with zero distortion. Monochromatic deep crimson red color grading, golden-hour backlighting between the two figures. High contrast chiaroscuro lighting, dramatic silhouettes. Gritty graphic novel poster style with subtle halftone texture. Cinematic, romantic yet powerful tone, film-poster quality. Stylized illustration, not photorealistic.
What makes it different: golden-hour backlighting softens the otherwise harsh red-black palette, and it’s the only prompt written for two subjects instead of one.
Negative Prompt (Use With Any Still-Image Variant)
Paste this after any of the portrait prompts above to stop the AI from drifting toward anime, cartoon, or overly smooth “beauty filter” results:
anime, cartoon, manga, chibi, comic outlines, thick line art, soft pastel colors, neon tones, flat lighting, low contrast, cute or cheerful mood, photorealistic skin, beauty retouching, bad anatomy, awkward pose, extra limbs, distorted body, text, watermark, logo
How to Use These Prompts
- Choose a clear, well-lit reference photo. Front-facing or three-quarter shots preserve identity best.
- Open Google Gemini, ChatGPT (DALL-E 3), or Midjourney and upload the photo.
- Paste one prompt variant from above, plus the negative prompt if your tool supports one.
- Add “strictly preserve facial identity, zero distortion” at the start of the prompt if the first result doesn’t look like you.
- For video, feed the finished still image into Kling AI, Luma Dream Machine, or Google Veo to add subtle hair and smoke motion, then sync the reveal to a beat drop in CapCut or VN.
Recommended Settings
| Setting | Recommended Value |
|---|---|
| Aspect ratio (photo) | 16:9 for posters, 9:16 for Reels/TikTok |
| Aspect ratio (video) | 16:9 or 2.39:1 widescreen |
| CFG / prompt strength | 7–9 |
| Image/denoise strength | 0.60–0.75 |
| Video duration | 3–6 seconds |
| Video motion strength | High |
FAQ
What does “AA23” actually mean?
It refers to Allu Arjun’s 23rd film, directed by Lokesh Kanagaraj. The teaser’s red-and-black visual identity inspired the AI prompt trend, though the prompts themselves are unofficial fan creations, not studio material.
Which AI tool gives the best AA23 results?
Google Gemini (Nano Banana model) is fastest for photo edits and free to start. Midjourney v6.1 gives the most polished, poster-quality texture. Stable Diffusion with a face LoRA gives the most consistent likeness across many generations.
Why doesn’t the AI-generated image look like me?
Add “strictly preserve facial identity” and “zero distortion” at the very start of your prompt, and lower the denoise/image strength to around 0.6 so the model changes the environment without redrawing your face.
Can I change the color from red to something else?
Yes. Swap “deep crimson red” for “deep emerald green,” “cobalt blue,” or “electric purple” while keeping the “neo-noir,” “chiaroscuro,” and “high contrast” language — that’s what carries the style, not the color itself.
What music is used in AA23 Reels?
Most viral versions use the “23 Theme” composed by Anirudh Ravichander, searchable on Instagram, Spotify, or YouTube as “23 Theme Anirudh.”
Does this work for both men and women?
Yes — the prompts are gender-neutral. For a feminine variation, add descriptors like “saree elegance” or “powerful female lead” to the pose description.
Is this free to use?
Yes, all the prompts on this page are open templates. Free tiers of Gemini or ChatGPT handle still images; high-resolution video generation may require a paid plan.
Hari Krishna is a video editor, designer, and AI workflow creator based in Kerala, India, with 9+ years in video editing and graphic design. He writes and tests every prompt, preset, and LUT published on Ri Imprint.