A new prompt structure for generating images of young female influencers in various settings and poses, tested with Nano Banana. The prompt includes detailed constraints for lighting, histogram, framing, and a list of forbidden elements to ensure tasteful and high-quality output.
2x2 grid, 16:9, do this for 4 young female influencers in very different settings and heart stealing poses data Coverage = Tasteful { secure :: Bool, opaque :: Bool } -- no explicit constructor exists (E₂) data Light = L { key :: Direction, fill_ev :: Float, neg_fill :: Direction, polarizer :: Bool } legal_light :: Light -> Bool legal_light l = key l == CamRight && fill_ev l == (-1.0) && neg_fill l == CamLeft data Histogram = H { skin_peak :: Float, shadow_floor :: Float } in_bounds :: Histogram -> Bool in_bounds h = skin_peak h <= 0.92 && 0.04 <= shadow_floor h && shadow_floor h <= 0.06 -- E₃ data Frame = F { aspect :: Ratio, verticals :: Bool, headroom :: Float, cut_joints :: Bool } legal_frame :: Frame -> Bool legal_frame f = aspect f == one_to_one && verticals f && headroom f ≈ 0.05 && not (cut_joints f) -- E₄ forbidden :: [Regex] -- the negative list, zero‑tolerance forbidden = [ /underage/ , /nudity|see_through/ , /explicit_pose|voyeur_angle/ , /pornographic_framing/ , /blur|low_res/ , /over_smoothing/ , /extra_limbs|bad_hands/ , /text|watermark|logo/ , /crushed_blacks/ , /tilted_verticals/ , /cut_joints/ ] render :: Subject -> Setting -> Photo render s st = let k = infer s st in assert (legal_light (light k) && in_bounds (hist k) && legal_frame (frame k)) $ editorial k
Click the copy button above to copy the prompt to your clipboard.
Open your preferred image generation tool like Midjourney, DALL-E, or Flux.
Paste the prompt into your tool, adjust settings if needed, and generate.
Try our AI Studio with this prompt or create something entirely new.