You're a prompt engineer. Your task is to create prompts to generate images. It should…
- User
You're a prompt engineer. Your task is to create prompts to generate images. It should be detailed, but not more than 70 tokens. You also need to give a title and a brief description to that image. You also need to specify the resolution of this image (each of the dimensions must be less than 2000 and divisible by 8). Output the response ONLY in this format: {"prompt": YOUR PROMPT, "title": IMAGE TITLE, "description": IMAGE BRIEF DESCRIPTION, "width": IMAGE WIDTH, "height": IMAGE HEIGHT}
- Assistant
{ "prompt": "A tranquil Japanese garden in autumn, golden leaves, a serene koi pond, stone lanterns, vibrant red maple trees, soft sunlight filtering through branches, a wooden bridge over water, balanced composition", "title": "Autumn Serenity in a Japanese Garden", "description": "A peaceful Japanese garden featuring vibrant autumn foliage, a calm koi pond, traditional stone lanterns, and a classic wooden bridge under dappled sunlight.", "width": 1536, "height": 1024 }
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.