原文
GPT Image 2 · en
Based on the understanding of the "{argument name="character name" default="character name"}" generate a unique and fully developed worldview, including scene style, cultural background, and spatial design. The composition should have an epic scale and cinematic framing. Feature the "{argument name="defend-type character" default="defend-type character"}" as the main protagonist in a high-quality anime art style, with clear facial features and natural body proportions (not chibi). The character should be shown in a full-body composition, standing slightly forward at the center of the frame, with a confident runway-like presence and posture. The character’s outfit style, materials, and overall color palette should be automatically derived from the generated worldview, featuring a premium design aesthetic with cohesive color layering. Position an exclusive "{argument name="guardian spirit beast" default="guardian spirit beast"}" behind the character on the left side. The AI should automatically generate the beast so it fully aligns with both the character and the worldview. The beast must be enormous, far larger than the character, with a structurally believable design. It should appear as a hybrid of physical form and energy, with intricate textures and layered lighting effects. The guardian occupies the upper portion of the composition, creating an overwhelming sense of pressure while advancing alongside the character. The character, guardian beast, and environment must share a unified artistic style while maintaining clear color contrast to avoid visual blending. Lighting should adapt naturally to the worldview while remaining bright, clean, and cinematic with high-end animated film quality. The overall aesthetic should resemble epic anime art combined with premium AAA game key visuals, featuring ultra-high detail, high resolution, and a clean composition. Negative requirements: no additional characters, no multiple spirit beasts, no chibi or low-quality cartoon styles, no text or watermarks, no blurry faces, and no deformed anatomy or structures.使い方
- パラメータを含む元のプロンプト全体をコピーします。
- 被写体、文字、構図を自分の要件に置き換えます。
- 対応する外部ツールで使用し、利用条件と結果を確認します。
開始前の確認
- 参考画像の要否を確認してください。互換モデルの表示は、そのモデルでの生成実績を意味しません。
- 名前、数値、画像内の文字を確認します。モデルの版で結果は変わります。
