Immersive Cultivation Novel Livestream Prompt

Original document

GPT Image 2 · en

Generate a 9:16 portrait screenshot of a Douyin live stream in a realistic style. The theme is {argument name="livestream theme" default="Three characters from the novel "A Record of a Mortal's Journey to Immortality" are live-streaming the sale of cultivation artifacts."} The image features three figures, whose demeanor, from left to right, is inspired by Han Li, Zi Ling, and Xuan Gu. Please do not display character names or use an anime style; they should resemble realistic cultivation-themed streamers. Clothing, makeup, and overall demeanor must retain an Eastern cultivation style while adhering to realistic photography principles. The product currently displayed is {argument name="displayed item" default="Golden Thunder Bamboo"} Place it in a prominent, central position in the foreground. The product card must clearly display "Golden Thunder Bamboo" and {argument name="item price" default=""9999 Spirit Stones""} The Golden Thunder Bamboo should possess the texture of a rare treasure, with clear details and a subtle yet realistic lightning halo. The live stream atmosphere should be lively and vibrant, including a realistic UI, bullet comments, likes, and viewer counts. Bullet comments should include relevant character nicknames similar to those used in real fan interactions. The overall effect must resemble a screenshot from a popular Chinese live stream—authentic, lively, and with a mobile phone-shot feel. Anime illustrations, movie posters, or game art are prohibited. English UI or exaggerated magical effects are also prohibited.

How to use

1. Copy the complete original prompt, including any parameters.
2. Replace the subject, text and composition with your own requirements.
3. Use a compatible external tool; check its current access conditions and review the result.

Before you start

• Check whether a reference image is required. A compatible model label does not mean this example was generated with that model.
• Keep names, figures and visible text accurate; results can vary between model versions.