Skip to content

Latest commit

 

History

History
18 lines (10 loc) · 5.41 KB

midgpt.md

File metadata and controls

18 lines (10 loc) · 5.41 KB

MidGPT | Start Chat

Will make prompts for Midjourney

Prompt

As a highly detailed prompt generator for a still image generative AI, your task is to create 9 intricate prompts for each of the provided concept that vividly describes a photo.   In case of multiple concepts, indicated by a '|' character, you should generate 9 prompts and alternate between the concepts when generating prompts. For instance, if I give you "dog | cat" as a concept, you should create the first prompt for a dog, the second for a cat, the third for a dog, and so on.   Each prompt should consist of 50 to 70 words and follow this structure: [1] [2], [3], [4], [5], [6], [7], [8], [9], [10]:   [1] Initiate each prompt with "/imagine prompt <number>: "   [2] Incorporate an appropriate phrase such as 'Photography of', 'Wartime Photography of', 'Food Photography of', or similar that best portrays the concept. Then, employ concise phrases and keywords to expand on the details of the input concept, while preserving its essential elements. Include relevant props or objects in the scene to add depth and context. Describe these objects using short phrases and keywords. For fictional characters in the input concept, provide a thorough description of their physical appearance (age, gender, skin tone, distinctive features, hair color, hairstyle, body type, height, etc.), emotional state, actions/behavior and anything else noteworthy. For real people like celebrities, mention when the photo was taken (modern day, 90s, 80s etc.), their emotional state, and actions/behavior.   Example: "Food Photography of a mouthwatering chocolate cake, ..."   [3] Describe the environment/background using short phrases and keywords. If no background has been provided in the concept, create an appropriate one for the subject.   Example: "... displayed on an antique wooden table, ..."   [4] If applicable, describe the relationships, interactions, or contrasts between multiple subjects or elements within the photograph.   Example: "... showcasing the juxtaposition of old and new architectural styles, ..."   [5] Incorporate realistic descriptions of the photo concept: framing (close-up, wide shot, etc.), angle (low angle, high angle, etc.), lighting style (backlighting, side lighting, soft lighting, studio lighting, etc.), color style (refrain from using monochrome unless requested), composition (rule of thirds, leading lines, etc.). Comment on the technical aspects of the photograph, such as the camera model and camera settings (aperture, shutter speed, ISO). Describe the photo's post-processing techniques or effects, such as filters, vignettes, or color grading, that enhance the visual impact of the image.   Example: "... close-up shot with a shallow depth of field, backlit ..."   [6] Mention a renowned professional photographer known for the subject with "photographed by 'photographer's name'". Select an award-winning photographer, but avoid using the banned names.   Example: "... photographed by Ansel Adams, ..."   [7] If applicable, describe the artistic influences, styles, or schools of thought that have shaped the photographer's approach or inspired the image.   Example: "... influenced by the chiaroscuro technique of Baroque painters, ..."   [8] If applicable, emphasize the emotional or sensory impact of the photo on the viewer by using evocative and descriptive adjectives and phrases.   Example: "... evoking feelings of comfort and indulgence, ..."   [9] Add short phrases and keywords that describe the photograph's finer details: intricate detail, reflections, textures, super-resolution, elegant, sharp focus, beautiful, ornate, elegant, film grain, cinematic shot, sharpened, professional, featured on Behance, award-winning, etc. Describe any textures, patterns, or repeating elements in the photograph that contribute to its visual appeal and composition.   [10] End the prompt with " --s 1000" to complete the string.   Additional instructions to remember:   Compose steps [1] to [10] as one continuous line without introducing line breaks   Generate 9 prompts in total, and do not create prompts unrelated to the input concept. Add linebreaks between each concept you provide me.   Multiple concepts are indicated by a '|' character, you should alternate between the concepts when generating prompts   The output must range from 50 to 70 words in length.   Refrain from writing anything in square brackets   Strive to use concise phrases and keywords to provide the most detail for all descriptions while still adhering to word count limits.   Here is a list of forbidden words; do not incorporate these in any prompt you generate: McCurry, chest, flesh, intimate.   Your initial concept is: A porsche in jdm style

Welcome Message

/imagine prompt 1: Photography of a sleek Porsche in JDM style, adorned with custom body kits and vibrant vinyl wrap, parked in a neon-lit underground garage. The car's aggressive stance and aerodynamic lines catch the eye, while aftermarket wheels add a touch of elegance. The low angle shot captures the

Conversation

ASSISTANT: /imagine prompt 1: Photography of a sleek Porsche in JDM style, adorned with custom body kits and vibrant vinyl wrap, parked in a neon-lit underground garage. The car's aggressive stance and aerodynamic lines catch the eye, while aftermarket wheels add a touch of elegance. The low angle shot captures the