Freepik

    Tips for using Freepik’s AI Suite tools

    AI Image generator prompt tips:

    • Use concise, direct prompts; avoid lengthy descriptions.
    • Omit contradictory terms, e.g., 'light winter clothes'.
    • Refrain from specifying styles in prompts; select from available options.
    • Employ negative prompts for intricate details, bypassing main prompt inclusion.
    • Eliminate repetition and specifications on pixels, sizes, and quality—adjust these in settings.
    • Exclude final usage mentions.
    • Avoid using 'Prompt enhancer' when the prompt is already very detailed.
    • Correct specific details, such as faces or hands, using Retouch or Upscale with 'none' imagination.
    • Avoid using words that end in '-tly', unless it’s really necessary.
    • Avoid the word 'background', as it causes blurriness.

    Retouch tips:

    • With Brush, use a broader selection than the targeted area for changes. Prioritize editing smaller, focused areas.
    • For multiple similar elements that require changes, select them collectively for uniform modification.
    • To remove something in the image, select with our tool and leave the prompt blank.

    Sketch to image tips:

    • For characters in different locations, set 'imagination' to zero. Add a background, describe characters and settings in the prompt, and refine and add details with the 'Paint' tool.
    • Use Retouch for additional detail enhancement.

    Upscale tips:

    • Before upscaling, review and refine the automatically generated prompt to eliminate undesired elements.

    Custom character tips:

    • Include different angles and expressions: images with varied angles, backgrounds, lighting, and facial expressions make the character feel natural and dynamic.
    • When you’re generating an image with a Custom Character, incorporate defining features in your prompts, such as 'pink hair' or 'smiling expression', to emphasize the character traits you wish to highlight. If you desire a specific appearance occasionally—like a particular outfit—be mindful not to overuse it in your images or prompts.
    • If you enter a very detailed prompt that might affect the Custom Character appearance, you can tune this down to get better results.
    • If you want to generate an image with more than one Custom Character, you will need to generate them separately and then edit the images to put them together.

    Custom style tips:

    • Incorporate a diverse range of scenes, colors, and objects. For optimal quality, aim to upload images that are at least 2000 pixels in width or height.
    • Uploading more images will enhance the richness and flexibility of the style, while fewer images will focus on specific details.
    • If you want to create a Custom Style from a prompt, keep it short but descriptive, focusing on mood, color and texture. You can make changes to refine the style later on.

    AI video generator tips:

    • It’s always better to generate using an image as a reference, as generating using only prompts will result in a low quality video.
    • The lipsync feature delivers the best results when used with close-up, centered photos of people, rather than illustrations or animals.

    Audio tips:

    • Enter (pause 250ms) or (pause 0.5s) in your prompt to add pauses to the Sound Effects.

    Tips for generating and editing images

    I want to generate characters in different poses:

    You can use Custom character, including different angles and expressions.

    I want to generate images with two characters:

    You can create two different characters separately with Custom Character, then create one image with several people and one of the custom characters, and then use the tool Retouch to swap one of the faces with the other character.

    You can also add up to two people to your images with Custom Characters. Go to the AI Image Generator > Character > New Character > upload photos of two people together > Create character. Name the character @name-name.

    I want to change styles:

    You can use Custom style, incorporating a diverse range of scenes, colors, and objects.

    I want to generate an image with text:

    We recommend using our AI Assistant, or Mystic mode in the AI Image Generator tool.

    I want to design multiple images of the same character while ensuring their clothing remains consistent:

    You can activate seed and change the prompt:

    I want to crop an image:

    At the moment it’s only possible in Expand.

    I want to run multiple prompts:

    To create images with several prompts at the same time, you can use the syntax for example: a black cat|a red cat|a yellow cat

    Can't find an answer to your question?