Dark Chat UI With Failed 4-Image Collage Prompt

Goal: Create a dark-mode mobile chat interface screenshot showing an AI image-generation result that failed to make four separate images and instead produced one 2x2 collage.

Canvas: Square screenshot, approximately 768×768 px, black background with a cropped mobile-app UI feel. Leave large empty black space on the right side and below the result.

Layout: At the top center, place a dark gray rounded prompt bubble containing the Chinese command text “生成4张独立的图片” with a thick red underline beneath it, and a smaller line below reading “收起 ^”. Below and left, show a status label in Chinese reading “已停止思考 ›”. Under that, place one generated image card with rounded top corners, containing exactly 4 photo panels arranged in a 2×2 grid, separated by thin white grid lines. Add a small circular download button with a downward arrow overlapping the lower-right corner of the collage. At the bottom, show a dark rounded input bar spanning the width: a plus icon on the left, placeholder text “有问题，尽管问”, and on the right the mode text “极速⌄”, a microphone icon, and a white circular voice/equalizer button.

Generated image content: The single generated image card must contain exactly 4 panels: top-left, top-right, bottom-left, bottom-right. Each panel shows the same young woman posing in front of a brightly lit juice shop or drink stand at night, with warm storefront lighting, neon signs, colorful drink menu boards, and a shallow depth-of-field smartphone-photo look. She wears a pink sleeveless crop top and a white pleated mini skirt, with hair styled in two cute buns. Her face is obscured by a soft rectangular blur for privacy. In each panel she holds a different plastic drink cup with a straw, posing slightly differently: 1) top-left, holding a red drink with both hands near her chest; 2) top-right, leaning forward holding a yellow/orange drink near her face beside neon “JUICE” signage; 3) bottom-left, holding a large orange drink close to her mouth in front of a bright menu board; 4) bottom-right, holding a pale pink drink while standing near a glass display case. The subject should be {argument name="character description" default="a young woman with two hair buns, pink sleeveless crop top, white pleated mini skirt, face blurred"}, photographed at {argument name="location" default="a colorful night-time juice shop storefront"}.

Visual style: Make the overall screenshot look like a real captured phone UI, not a polished mockup. Use slightly compressed screenshot quality, soft glow from neon lights, realistic photo colors in the collage, and minimalist dark UI elements. The Chinese command in the top bubble should be {argument name="command text" default="生成4张独立的图片"}. The status label should be {argument name="status text" default="已停止思考 ›"}. The bottom input placeholder should be {argument name="input placeholder" default="有问题，尽管问"}.

Constraints: The result must show one combined 2×2 collage rather than four separate image cards. Include exactly 4 photo panels and exactly 1 download icon. Preserve the dark chat interface, Chinese UI text, red underline, and bottom input bar. Do not add extra panels, extra people, watermarks, or captions inside the generated photo collage.