We have simplified our image generation feature to allow you to create custom images that complement your search within seconds.
How does it work?
Add your prompt to the search input and try to be as specific as possible with your instructions. Perplexity will automatically generate the image (you no longer have to click an additional button for generating the image after prompting).
Here are some examples:
What model is used for image generation
The default model for image generation is GPT Image 1. We have added the latest image generation model from OpenAI that supports contextual image generation and editing. We will continue adding the latest models as they become available.
Can I choose what model is used for image generation?
Yes, can set your model preference for image generation from the settings menu. Click the “Settings” icon in the user, select “Preferences”, and then choose an “Image generation model” of your choice. The currently available options include GPT Image 1, Gemini 2.0 Flash Experimental, FLUX.1, and DALL-E 3.
Can I edit generated images?
If you're unhappy with the first result you can click on "Regenerate" below the image and we will automatically create a new image. Pro subscribers can regenerate an image unlimited times.
Alternatively, you can add a new prompt in the same Thread (make sure to include something like "generate an image" in your follow-up prompt).
How many images can I generate per day?
Image generation is available as an enhanced query. Non-subscribers get a limited amount of enhanced queries per day, while Pro subscribers receive unlimited enhanced queries.
Pro subscribers can use GPT Image 1 up to 150 times per month.
Can I use generated images for commercial purposes?
Images generated through our service are for personal, non-commercial use only. For more details, please refer to our Terms of Service.