Comparison of GPT Image 2 and Kunoichi

Objective: Create a side-by-side comparison image showcasing an anime female ninja illustration optimized with GPT Image 2, highlighting the improved details and cinematic rendering. Canvas: 16:9 wide white background, simple contrast layout. Two tall, vertical anime illustration panels are placed side-by-side: one on the left and one on the right, separated by a white column. Layout: Uses 5 main visible elements: one left illustration panel, one right illustration panel, one large gray arrow pointing to the right in the center, one bold black "GPT Image2" text label above the arrow, and one bold black Japanese title "描き込み量を\n极限まで上げて" below the arrow. The left panel occupies the left third of the canvas, the right panel occupies the right third, and the middle column contains text and arrows. Main Details: Depict the same high-quality, dynamic anime-style female ninja in both panels.{argument name="character name" default="Ai-chan"} She leaped through the air with dramatic movements. She had flowing hair {argument name="hair color" default="gold"} With a high ponytail, her face obscured by a square brown blur, she wears black fingerless gloves and a purple floral kimono-style female ninja outfit, complete with pink cherry blossom patterns, a short skirt, thigh-high straps, a waist pouch, and soft ninja boots. She wields three clearly visible weapons: two short swords crossed above her head and a curved sword slung across her waist. Her pose features one leg raised high and the other bent backward, conveying a strong sense of movement and her hair blowing in the wind. The left panel features a richly detailed yet slightly soft anime illustration against a backdrop of Japanese temple architecture at sunset. It includes stone steps, a red pagoda, curved temple eaves, stone lion statues, distant hazy mountains, a pink sky, cherry blossom branches, falling petals, and a faint pink spiral of magical energy surrounding the character. Right Panel: A more refined, sharper, more detailed, and cinematic version of the same scene, featuring richer contrast, denser cherry blossoms, more ornate roof carvings, stronger lighting, more saturated sunset colors, more petals, clearer stone textures, brighter pink energy vortexes, and additional environmental details. Maintains the same character poses and clothing as the left panel, but with a significantly more refined rendering. Visual Style: Ultra-high-quality cinematic anime art, extremely detailed lines, soft shadows, high detail, dramatic rim lighting, glossy highlights on blades, a cherry blossom pink atmosphere, fantastical action composition, intricate Japanese architecture, dynamic motion blur only on petals and hair, and a clear character focus. Text Content: The central text must contain two text blocks: above the arrow is "GPT Image2," and below the arrow are two lines of Japanese titles: the first line is "描き込み量を," and the second line is "极限まで上げて." Constraints: Retain the central white gap and outer white margin. Do not add any logos, watermarks, social media UIs, additional descriptive text, or extra panels. The faces in both panels should remain obscured by a square blur.