# Text-to-Image ### How It Works 1. **Choose your method:** * **Quick path:** Use the top-right model selector and type in chat * **Full control:** Enable Generate Media and configure all settings 2. **Select image model** - Choose from Flux, GPT Image, Nano Banana, Grok, etc. 3. **Set aspect ratio** - 16:9, 9:16, 1:1, 4:5, or custom 4. **Set resolution** - Based on model capabilities 5. **Write your prompt** - Describe the image you want 6. **Generate** - Click send and wait for your image ### Two Ways to Generate #### Quick Path (Top-Right Model Selector) **Best for:** Fast, simple image generation during conversation 1. **Set default image model** in the top-right selector 2. **Type in chat:** "Generate me an image of a sunset over mountains" 3. **Image appears** - No settings to configure 4. **Done** - Quick and simple **Characteristics:** * ✅ Fast and simple * ✅ Natural language requests * ❌ Limited control (no aspect ratio/resolution) * ❌ Uses default settings #### Full Control (Generate Media Mode) **Best for:** Production work, specific requirements 1. **Enable Generate Media** toggle 2. **Select image model** from the dropdown 3. **Set aspect ratio** (16:9, 9:16, 1:1, 4:5, etc.) 4. **Set resolution** (if available for model) 5. **Enter detailed prompt** 6. **Generate** with full control **Characteristics:** * ✅ Complete control over all settings * ✅ Custom aspect ratios and resolutions * ✅ Advanced model-specific options * ✅ Best for production work ### Writing Effective Prompts #### Essential Elements A good image prompt includes: 1. **Subject** - What or who is the main focus? 2. **Setting** - Where does this take place? 3. **Style** - What's the visual aesthetic? 4. **Lighting** - What's the lighting like? 5. **Mood/Atmosphere** - What's the feeling? 6. **Details** - Specific elements, colors, textures #### Prompt Structure ``` [SUBJECT] in [SETTING], [STYLE], [LIGHTING], [COLOR PALETTE], [MOOD], [DETAILS] ``` #### Good Examples ✅ **Detailed and Specific:** ``` A cozy coffee shop interior with warm wooden furniture, large windows streaming golden hour light, plants hanging from the ceiling, vintage decor, soft bokeh background, peaceful and inviting atmosphere, cinematic style, film grain texture ``` ✅ **Style-Focused:** ``` Futuristic cityscape at night, cyberpunk style, neon lights reflecting on wet streets, flying vehicles in the background, dramatic lighting, vibrant colors, high-tech atmosphere, detailed and sharp ``` ✅ **Mood-Driven:** ``` Serene mountain landscape at dawn, misty valleys, soft pastel colors, dreamy atmosphere, ethereal lighting, peaceful and tranquil mood, cinematic wide shot, film aesthetic ``` #### Bad Examples ❌ **Too Vague:** ``` A picture ``` ❌ **Missing Details:** ``` Coffee shop ``` ❌ **No Style:** ``` Person in a room ``` ### Model-Specific Tips #### Flux 2 Max **Best For:** Highest quality, detailed images **Prompt Tips:** * Use detailed descriptions * Mention specific styles and aesthetics * Describe lighting and atmosphere * Include texture and material details **Example:** ``` Professional product photo of a vintage camera on a wooden desk, studio lighting with soft shadows, shallow depth of field, cinematic composition, great detail, sharp focus ``` #### GPT Image 2 **Best For:** Canvas editor workflows, composition, high-resolution output **Prompt Tips:** * Works well with composition requests * Supports complex scene descriptions * Good for layered imagery and multi-image edits (up to 4 images) * Best when used with Canvas Editor * Supports 8 aspect ratios: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9 **Example:** ``` Modern minimalist workspace with a laptop, plants, and a coffee cup, clean composition, natural window light, professional photography style ``` #### Nano Banana Pro **Best For:** Fast, high-quality results **Prompt Tips:** * Balanced quality and speed * Works well with varied styles * Good for quick iterations * Supports multiple images (up to 14) **Example:** ``` Vibrant street art mural on a brick wall, urban setting, bright colors, graffiti style, dynamic composition, energetic atmosphere ``` #### Z-Image Turbo **Best For:** Speed, quick iterations **Prompt Tips:** * Fast generation * Good for previews and tests * Simpler prompts work fine * Quick turnaround **Example:** ``` Sunset over the ocean, warm colors, peaceful mood ``` #### Grok (xAI) **Best For:** Fast generation, mobile-first content **Prompt Tips:** * Natural language prompts work well * Simple, direct descriptions are effective * Great for social media and mobile content * Supports image-to-image editing as well **Unique Features:** * **Mobile aspect ratios** - 19.5:9, 9:19.5, 20:9, 9:20 for smartphones * **Panoramic formats** - 2:1, 1:2 for ultra-wide/tall content * **Resolution options** - 1K or 2K (with upscale) * **Fast turnaround** - Quick generation times **Example:** ``` A vibrant street food market at night, colorful neon signs, steam rising from food stalls, bustling atmosphere, urban energy ``` ### Aspect Ratio Selection #### Common Aspect Ratios | Aspect Ratio | Use Case | Dimensions (Example) | Models | | ------------ | ------------------------- | -------------------- | ------ | | **16:9** | Landscape, widescreen | 1920×1080, 1536×864 | All | | **9:16** | Portrait, mobile, social | 1080×1920, 864×1536 | All | | **1:1** | Square, Instagram | 1024×1024, 1536×1536 | All | | **4:5** | Portrait, Instagram posts | 1024×1280, 1536×1920 | Most | | **21:9** | Ultra-wide, cinematic | 2048×864 | Most | | **2:1** | Panoramic landscape | 2048×1024 | Grok | | **1:2** | Extra tall portrait | 1024×2048 | Grok | | **20:9** | Smartphone landscape | 2048×922 | Grok | | **19.5:9** | Modern phone landscape | 2048×945 | Grok | | **9:19.5** | Modern phone portrait | 945×2048 | Grok | | **9:20** | Extended phone portrait | 922×2048 | Grok | #### Choosing the Right Aspect Ratio **16:9 (Landscape):** * Desktop wallpapers * YouTube thumbnails * Website headers * Widescreen displays **9:16 (Portrait):** * Mobile wallpapers * Instagram Stories * TikTok content * Phone screens **1:1 (Square):** * Instagram posts * Profile pictures * Social media * Balanced compositions **4:5 (Portrait):** * Instagram feed posts * Portrait orientation * Social media * Vertical displays ### Resolution Settings #### Model-Specific Resolutions | Model | Max Resolution | Available Resolutions | | ------------------- | ------------------------------- | ----------------------------------------------------------------- | | **Flux 2 Max** | 1536×1536 | Up to 1536px on longest side | | **GPT Image 2** | 1792×1792 | 8 aspect ratios; any dimension (multiples of 16, max edge 3840px) | | **Seedream v5** | 2048px native (4K with upscale) | 10 aspect ratios; 2K native, 4K via Crisp Upscale (max 3072×3072) | | **Nano Banana Pro** | 1536×1536 | Up to 1536px on longest side | | **Grok** | 2K (upscaled) | 1K or 2K (with upscale) | | **Z-Image Turbo** | 1024×1024 | Up to 1024px on longest side | #### Resolution Guidelines **Lower Resolution (720p-1024p):** * Faster generation * Lower cost * Good for previews * Web use **Higher Resolution (1536p+):** * Better quality * Print-ready * Professional work * Slower generation ### Tips for Best Results 1. **Be specific** - "Cozy coffee shop with warm lighting" vs. "coffee shop" 2. **Include style** - "Cinematic", "photorealistic", "illustration style." 3. **Describe lighting** - "Golden hour", "soft natural light", "dramatic shadows." 4. **Mention mood** - "Peaceful", "energetic", "mysterious." 5. **Add details** - Textures, colors, composition elements 6. **Use appropriate model** - Flux for quality, Turbo for speed 7. **Match aspect ratio to use case** - 16:9 for landscape, 9:16 for portrait ### Common Use Cases #### Product Photography **Example:** * Prompt: "Professional product photo of \[product] on white background, studio lighting, clean composition, high detail." * Aspect ratio: 1:1 or 16:9 * Model: Flux 2 Max or GPT Image 2 #### Social Media Content **Example:** * Prompt: "Vibrant social media post design, \[theme], modern style, eye-catching colors, engaging composition." * Aspect ratio: 1:1 (Instagram) or 9:16 (Stories) * Model: Nano Banana Pro or Z-Image Turbo #### Concept Art **Example:** * Prompt: "Concept art of \[subject], \[style], detailed environment, atmospheric lighting, creative composition." * Aspect ratio: 16:9 or 21:9 * Model: Flux 2 Max #### Backgrounds **Example:** * Prompt: "Abstract background, \[colors], \[texture], seamless pattern, versatile design." * Aspect ratio: 16:9 or custom * Model: Any model ### Troubleshooting #### "Image doesn't match my prompt." **Solutions:** * Be more specific in your description * Include style keywords (cinematic, photorealistic, etc.) * Add more details about composition and lighting * Try a different model (Flux for better understanding) #### "Wrong aspect ratio" **Solutions:** * Check your aspect ratio setting before generating * GPT Image 2 supports 8 aspect ratios (1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, 21:9) * Verify the aspect ratio matches your use case * Regenerate with the correct aspect ratio #### "Quality is too low." **Solutions:** * Use higher resolution settings * Switch to Flux 2 Max for the best quality * Ensure your prompt is detailed enough * Check model capabilities for max resolution #### "Generation is too slow." **Solutions:** * Use Z-Image Turbo for faster generation * Lower resolution settings * Use Nano Banana Pro for balanced speed/quality * Check your internet connection *** **Next:** Learn about [Image-to-Image ](/features/image-generation/image-to-image.md)for editing existing images. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.chatvideopro.com/features/image-generation/text-to-image.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.