A good Midjourney prompt clearly describes the image you want to create. Focus on the visual elements rather than giving instructions. Include details such as the subject, environment, lighting, style, mood, colours, and composition. Midjourney works best when prompts are descriptive and visually focused rather than conversational. Experimenting with different keywords and parameters can dramatically change the final result.

"determined female detective, standing in a rain-soaked neon city street, cinematic lighting, cyberpunk atmosphere, dramatic shadows, shallow depth of field, ultra realistic photography"

"young entrepreneur holding futuristic AI hologram, vibrant colours, high contrast lighting, excited expression, professional YouTube thumbnail style, sharp focus, modern technology background"

"Texas ranch at sunrise, cattle grazing across rolling hills, golden hour lighting, cinematic landscape photography, realistic detail, National Geographic style"

"teenage girl walking along abandoned train tracks at sunset, dusty southern town in the distance, cinematic storytelling, warm golden light, emotional atmosphere, film still"

"ancient floating city above the clouds, epic fantasy architecture, dramatic sunlight breaking through storm clouds, highly detailed concept art, breathtaking scale --ar 16:9"

❌ Writing long instructions as if speaking to ChatGPT.

✅ Describe what you want to see visually.

"elderly fisherman on a wooden dock, misty morning, soft lighting, realistic photography"

❌ Leaving out important visual details.

✅ Include key elements such as:

Subject

Environment

Lighting

Style

Mood

Composition

Use descriptive keywords rather than full instructions.

Start with the main subject, then add environment and style details.

Specify lighting such as golden hour, neon, cinematic, studio, or dramatic.

Include camera and composition terms such as close-up, wide shot, aerial view, or shallow depth of field.

Think like a cinematographer—describe the scene as if you were directing a camera.

Experiment with multiple variations and refine prompts based on the results.

For AI filmmaking, create key scenes, storyboards, character concepts, environments, and YouTube thumbnails before moving into video generation.

A good Nano Banana prompt is clear, structured, and highly descriptive, combining both what you want to see and how you want it to look. Unlike conversational AI prompts, image prompts should focus on visual specificity, composition, style, and output constraints. The more clearly you define the subject, setting, and aesthetic direction, the more controlled and consistent your results will be.

Nano Banana performs best when prompts include subject, context, style, camera or framing cues, 

lighting, and purpose (e.g., thumbnail, cinematic still, concept art). It also benefits from structured prompts that separate ideas clearly.

"Create a cinematic film still of a lone astronaut standing on a red Martian landscape. Wide shot composition, dramatic lighting with long shadows, dust in the air, realistic sci-fi tone, emotional mood of isolation, ultra-detailed."

"Design a high-impact YouTube thumbnail showing a creator reacting to AI-generated visuals. Bright contrasting colors, expressive face, futuristic holograms in the background, bold composition with clear focal point, optimized for engagement."

"Generate a storyboard frame of a teenage girl running through a rainy neon-lit city at night. Low-angle camera shot, motion blur, cinematic lighting, cyberpunk aesthetic, film still style."

"Create an illustration showing how AI video generation works. Clean, minimal infographic style, labelled layers showing text input, model processing, and video output. Modern tech aesthetic, easy to understand."

"Depict a floating medieval city above clouds at sunrise. Epic scale, soft golden lighting, highly detailed architecture, fantasy concept art style, atmospheric depth, cinematic composition."

❌ Writing vague or abstract prompts
"Make something cool with AI in it"

✅ Be visually specific
"Futuristic laboratory with glowing holographic screens, scientist interacting with floating data visualisations, cinematic lighting"

❌ Forgetting purpose or format
✅ Always define output intent:

• YouTube thumbnail 
• cinematic still 
• storyboard frame 
• concept art 
• educational diagram 

❌ Overloading with unrelated ideas
❌ Mixing too many styles or scenes in one prompt
✅ Keep one clear visual concept per prompt

Use clear visual structure: subject → environment → action → style → lighting → composition.

Think like a film director or art director, not a writer.

Specify the intended use (thumbnail, storyboard, cinematic shot, etc.).

Include camera language such as wide shot, close-up, low angle, depth of field, motion blur.

Define visual style clearly (cinematic, photorealistic, illustration, concept art, infographic).

Keep prompts focused—one strong idea per image works better than multiple ideas competing.

Iterate often: small prompt changes can significantly improve output quality.

A good Kling AI image prompt is structured, detailed, and visually descriptive, focusing on how the image should look rather than how you want the AI to behave. Kling works best when prompts clearly define the subject, environment, lighting, style, composition, and any transformations or edits.

Unlike conversational prompting, Kling responds most effectively to formula-based inputs, where each part of the scene is explicitly described. The more precise and structured your prompt is, the more control you have over the final image.

Kling also supports editing and reference-based prompting, meaning you can modify existing images by specifying exactly what should change while keeping everything else consistent.

A strong Kling prompt typically follows this structure:

Subject + Environment + Action + Style + Lighting + Composition + Details/Modifiers

For editing tasks, the structure shifts slightly:

Keep X unchanged → modify Y → desired result

For face or character consistency:

Keep facial features + define hair + clothing + action + background + lighting + mood

"Young woman standing on a rooftop overlooking a neon city at night, rain falling, reflective surfaces, cinematic lighting, shallow depth of field, ultra-realistic style, film still composition, dramatic mood"

"Minimalist product shot of a glass perfume bottle on polished black marble surface, soft studio lighting from the left, subtle reflections, luxury advertising style, high detail, shallow depth of field"

"Close-up portrait of a weary astronaut inside a spacecraft, helmet removed, soft blue control panel lighting, realistic skin texture, emotional expression, cinematic sci-fi style"

"Keep everything else unchanged, change the background from a forest to a futuristic city skyline at night, add neon lighting reflections, maintain subject position and pose"

"Keep facial features of the character, long dark hair, wearing a red jacket, standing in a snowy street, soft evening lighting, calm atmosphere, cinematic realism"

❌ Writing vague prompts
"Make a cool futuristic image"

✅ Be visually specific
"Futuristic city at night, flying vehicles, neon signs, rain-soaked streets, cinematic lighting, wide-angle view"

❌ Treating Kling like a chatbot
❌ Using instructions instead of visual descriptions
❌ Overloading the prompt with unrelated ideas

❌ Forgetting structure in edits
❌ Not specifying what should stay the same
✅ Always use:

• Keep everything else unchanged 
• Specify exact modification 
• Describe final visual result 
• 
  

Build prompts like a film director describing a shot, not a writer giving instructions.

Always define:

• Subject (who or what is in the image) 
• Environment (where it is happening) 
• Lighting (golden hour, neon, studio, natural light) 
• Style (cinematic, photorealistic, editorial, concept art) 
• Composition (close-up, wide shot, centered, low angle) 

For edits, clearly separate:

• what stays the same 
• what changes 
• what the final result should look like 

Use short, structured phrases instead of long sentences for more consistent results.

Experiment with variations—small prompt changes in Kling can significantly change mood, realism, and composition.

A good DALL·E prompt uses clear, natural language to describe exactly what you want to see. DALL·E excels at understanding detailed scene descriptions, making it ideal for concept art, storyboards, illustrations, and photorealistic images.

The more specific you are about the subject, setting, style, lighting, and composition, the more accurate the results will be.

Subject + Environment + Style + Lighting + Composition + Details

Cinematic Scene

"A lone cowboy walking through a dusty desert town at sunset, cinematic lighting, wide-angle shot, realistic photography, dramatic atmosphere."

YouTube Thumbnail

"Excited creator surrounded by AI-generated images, bright colours, high contrast, professional YouTube thumbnail style."

❌ Being too vague

"Create something futuristic."

✅ Be specific

"Futuristic city skyline at night with flying vehicles, neon lights, cinematic realism."

• Write naturally, as if describing a scene to an artist. 
• Include style, mood, and lighting. 
• Specify camera angles when relevant. 
• Refine prompts through multiple iterations. 
• 
  

A good Firefly prompt focuses on the visual outcome and uses descriptive language similar to how a designer or photographer would describe a project. Firefly performs particularly well when prompts clearly define the subject, environment, and artistic style.

Subject + Action + Environment + Style + Lighting

Advertising Visual

"Luxury wristwatch on black marble, soft studio lighting, premium product photography, sharp focus."

Documentary Image

"Rancher working at sunrise, golden hour lighting, realistic photography, National Geographic style."

❌ Using vague instructions

"Make a cool image."

✅ Describe what should appear in the image

"Modern office with AI-powered workstations, natural light, realistic photography."

• Think like a designer or photographer. 
• Use simple, descriptive language. 
• Define style and lighting clearly. 
• Focus on the visual result rather than instructions. 
• 
  

A good Grok image prompt combines clear visual descriptions with conversational refinement. Grok works well when you describe the scene clearly and then iterate through follow-up prompts to improve results.

Subject + Setting + Style + Mood + Key Details

Concept Art

"Ancient floating city above the clouds, fantasy architecture, golden sunrise, cinematic concept art."

Film Poster

"Young hacker standing in a neon-lit alley, cyberpunk atmosphere, dramatic lighting, movie poster style."

❌ Overcomplicating prompts

❌ Combining unrelated ideas

✅ Focus on one strong visual concept

• Start simple and refine. 
• Describe mood and atmosphere. 
• Use follow-up prompts to adjust results. 
• Think visually rather than technically. 
• 
  

A good Meta AI prompt uses clear, conversational language and focuses on the main visual elements of the image. Meta AI is designed to be approachable and works best when prompts are direct and easy to understand.

Subject + Environment + Style + Mood

Travel Scene

"Mountain lake at sunrise, calm water reflections, vibrant colours, realistic photography."

Character Portrait

"Young filmmaker holding a camera, warm lighting, modern city background, cinematic portrait."

❌ Leaving out important details

"Generate a landscape."

✅ Add context and style

"Snow-covered mountain landscape at sunrise, realistic photography, dramatic lighting."

• Use natural language. 
• Be specific about the scene. 
• Include style and mood. 
• Refine prompts through conversation. 
• 
  

A good Seedream prompt is highly structured and visually detailed. Seedream excels when prompts precisely define the subject, composition, lighting, and artistic style, making it ideal for cinematic and photorealistic imagery.

Subject + Environment + Composition + Lighting + Style + Details

Cinematic Film Still

"Teenage girl walking along train tracks at sunset, small southern town in the distance, cinematic lighting, shallow depth of field, film still."

Photorealistic Portrait

"Close-up portrait of an astronaut, realistic skin textures, soft blue lighting, ultra-detailed photography."

❌ Being too general

"Create a portrait."

✅ Define every important visual element

"Close-up portrait of a female scientist, laboratory background, soft studio lighting, realistic photography."

• Use structured descriptions. 
• Focus on visual precision. 
• Include composition and lighting. 
• Separate subject, style, and environment clearly.