You can generate images directly inside Perplexity Pro using text prompts. This feature lets you create custom visuals for presentations, social media, or brainstorming without leaving the chat window. Perplexity Pro uses advanced AI models to turn your descriptions into high-resolution images. This article explains how to access the image generation tool, write effective prompts, and adjust output settings for the best results.
Key Takeaways: How to Generate Images With Perplexity Pro
- Pro subscription required: Image generation is available only in Perplexity Pro, not the free tier.
- Type a prompt after selecting the image model: Choose the image-focused model from the drop-down menu before typing your description.
- Use descriptive keywords: Include style, subject, colors, and composition details to get accurate results.
How Image Generation Works in Perplexity Pro
Perplexity Pro integrates with AI image generation models that create images from text descriptions. The feature is built into the standard chat interface. When you select an image-capable model, the system interprets your prompt and generates a new image based on the instructions. You do not need to install any additional software or plugins. The image is generated within seconds and appears directly in the conversation thread.
Before you start, confirm that you have an active Perplexity Pro subscription. You can check this by clicking your profile picture and selecting Subscription. The free plan does not include image generation. Also ensure that you are using a supported web browser such as Chrome, Edge, Firefox, or Safari. Mobile apps for iOS and Android also support image generation with the same steps.
Steps to Generate an Image Using Perplexity Pro
Follow these steps to create an image from scratch. The process works identically on the web app and mobile apps.
- Open a new conversation
Click the New Thread button in the left sidebar. This clears any previous context and gives you a fresh prompt field. - Select the image generation model
Click the model selector drop-down at the top of the prompt area. Choose an option labeled with Image generation support. Common labels include Pro Image or Default with Image. If you do not see this option, your account may not have Pro access. - Write your image prompt
Type a detailed description of the image you want. Include the subject, style, colors, lighting, and composition. For example: A photorealistic red fox standing in a snowy forest at sunset, with soft golden light filtering through pine trees. Avoid vague words like nice or good. - Send the prompt
Press Enter or click the send arrow. Perplexity processes the request and displays the generated image in the conversation. The image appears as a large preview with a download button below it. - Refine the result if needed
If the image does not match your expectations, type a follow-up prompt. For example: Make the fox smaller and add more snow on the ground. The model generates a new image based on the updated instructions. - Download the image
Click the download icon under the image to save it to your device. The file is saved as a PNG image at the original resolution generated by the model.
Common Mistakes and How to Avoid Them
Prompt is too short or too vague
Short prompts like a cat or a car produce generic results. The AI does not know what style, background, or mood you want. Always include at least five descriptive elements: subject, action, setting, lighting, and color palette. Example: a black cat sleeping on a red velvet couch in a dimly lit Victorian library, warm amber lighting.
Selected model does not support images
Some Perplexity models are text-only. If you select a model without image capability, the AI responds with text instead of an image. Always check the model label before typing your prompt. The model drop-down shows a small camera icon or the word Image next to supported models.
Asking for copyrighted or trademarked content
The image generation model may refuse prompts that request images of specific branded characters, logos, or celebrities. If your prompt is rejected, rephrase it to describe the concept without naming the protected entity. For example, instead of Mickey Mouse, write cheerful cartoon mouse with round ears and red shorts.
Expecting instant high resolution
The generated image resolution depends on the model. Some models produce images up to 1024×1024 pixels. If you need a larger or higher-resolution image, use a follow-up prompt like Increase resolution to 2048×2048 pixels. Not all models support this, so results may vary.
Perplexity Pro Image Generation vs Other AI Image Tools
| Item | Perplexity Pro Image Generation | Standalone AI Image Tools (e.g., Midjourney, DALL-E) |
|---|---|---|
| Access method | Built into chat interface | Separate website or app |
| Subscription cost | Included with Perplexity Pro | Requires separate monthly payment |
| Maximum image resolution | Up to 1024×1024 (varies by model) | Up to 1792×1024 or higher |
| Style control | Text prompt only | Text prompt plus parameters (aspect ratio, style presets) |
| Image editing after generation | No built-in editor | Often includes inpainting, outpainting, or variations |
Perplexity Pro image generation is best for quick, simple visuals created within a research or chat workflow. Standalone tools offer more advanced editing and higher resolution for professional design projects.
You can now generate images inside Perplexity Pro by selecting the correct model and writing detailed prompts. Start with a clear description of your subject, style, and setting to get the best results. For more control, refine your prompt with follow-up instructions until the image matches your vision.