Generate Image with Prompts

Prompt

Keywords that describe the image you want to make. It could be phrases or words. Separate with comma.

Describe with Tags

PixAI's models work best when you word your prompts in the form of descriptive 'tags' or short phrases. You'd want to separate your tags with commas. Here's a basic example of how tags are worded:

Autocomplete

If you enable autocomplete, the AI will assist you in completing your prompts.

###### Emphasis
Tell the AI what tags you wish it to focus on by surround tags with '()' --or in multiple: '(())'-- to strengthen it. Or you could do so by weakening a tag with '[]'.


The order of your prompt matters: Try to describe the main object/character at the beginning of your prompt. For example, for character portraits, our users like to start with '1girl' or '1boy' as their first tag.
Ratios: The usage of ratios or weights with our models does not always produce the most stable or obvious results. Please keep this in mind while experimenting with your prompts.

Notes

In PixAI, you can use the Note function to save the setting template which can memorise prompts, negative prompts, models, size, sampling steps and methods, base image. Once the note is saved, you can use it as a template any time by importing it.



You can save the current settings or create a new one.



You’ll need to enter a title for your current settings to be saved as a note.



Or, you can create a new note and input everything you’d like to be remembered.

Natural language

Use the Prompt Helper to describe the image you'd like to create with natural language! This means you can write full sentences rather than separating short phrases into prompts. We support a wide range of languages, so you don’t need a translator to convert your language into English for the AI to understand your meaning.

Image to Prompt

This feature can be used to get the prompts from an uploaded image in just a few seconds!



After uploading, the prompts will appear on the right, and you can choose from them.



Click "Confirm," and the selected prompt will be applied.

Generate Image with Base Image

Image to Image

Upload Image

Upload an image as a reference, and the generated results will closely resemble the uploaded reference image.

Paint

You can use the Paint feature to modify specific parts of the reference image. For example, if you use the brush to cover the character's eyes and type "blue eyes" in the prompt area, you'll receive an image that is nearly identical to the original but features blue eyes.

Image Edit

The Edit mode allows you to modify your generated images as you wish. You can import your generated results, upload an image from your local device, or simply paste an image from your clipboard to begin editing.



💡Note:
You can use scroll up and down to control the whole canvas size.
Switch mode between Inpaint and Outpaint will discard all changes.

InPaint

The Inpaint mode is similar to the Paint feature in the Reference Image. You can use the brush to cover parts you want to modify



or use the box to select areas for changes. Don’t forget to enter prompts that describe the changes you envision!

OutPaint

Click the Outpaint button to switch to Outpaint mode.





You can stretch the box to extend the picture. The Outpaint mode can be used to add another character or other elements to the image.

Add Booster

Enhance your creations effortlessly with these convenient features in the Add Booster section!

Face Fix

Add Face Fix for flawless facial details, ensuring lifelike and polished results.

HiRes

HiRes can be used to boost image resolution for sharper and more detailed outputs.



Note: Enabling HiRes allows you to generate only one image at a time instead of a batch.

Quality Tag

Join the PixAI membership to customize your quality tag and add preset quality tags for different models! This feature automatically adds a series of quality tags to submitted prompts, ensuring high-quality image generation.

When uploading a model, you will have the option to set 1-3 default quality tags, allowing everyone using it to generate images that meet their expectations.

Style Mix

This feature lets you create images in various styles, bringing excitement and randomness to your creations!



Comparison of generation results of different art styles based on the same prompt and model↓





Note: This booster is only available for model Waterfront 0.9 and 1.0.

Generation Result

Animate

You can animate your images with just one click!

Note: This Professional Mode of the animation feature is the same as the image-to-video feature.

Describe

Same as Image to Prompt. You can get generated prompts from selected task.

Delete

You can delete your tasks on the generation panel.

Variation

Variation will take the current image as base image to generate 3 new different images using the same prompts and settings with the base image task.
💡Note: This feature cannot be used for upscaled artworks, including those generated from an edited base image.

Edit

Edit will bring the task to the "Edit" page.

Use as Base Image

Take the selected image as reference image.

Download

Download this image as a PNG file.

Enhance

You can choose to upscale or enhance an image to improve its quality. Upscaling will increase the size of your image without altering it, resulting in a higher resolution version. Enhancing the image will generate a higher quality version based on the original.

Publish

You can publish your task to PixAI community to earn extra credits once you fill in the post details.

Favorite

Add the tasks you like to your favorites for quick access!

Settings

Model

Your artwork model. Different models have different styles, such as photo-like, realistic, anime and so on.



You can discover more models in our Model/LoRA market by clicking "Market." This will redirect you to the Model Market page, where you can search for models using the available filters.



You can manage your models if you're a PixAI member. Click on "Bookmark" to switch to the model lists you have bookmarked as your preset model lists.

LoRA

Expanded learning packet for model. It can make generation more targeted or detailed on specific character/style. It can let AI be able to recognize specific prompts.
A LoRA can make AI recognize specific prompts for a better result.
You can find more LoRAs in PixAI Model Market

Size

You can select from a list of preset image dimensions or enter a custom size for your image. Become a PixAI member to unlock larger size limits!

Number of Images

You can choose to generate a batch of 4 images or just one image at a time.

Note: If you select "hires" from the boosters, you will only be able to generate one image at a time.

ControlNet

ControlNet can analyse the image you upload to shape the frame of your generation.

Composition

Composition allows you to create images by specifying details about what content will appear in different parts of the canvas.

Note: This feature is not available for SDXL models.

Advanced

Negative prompts

Negative prompts tell the AI what to avoid. On our generation page, we provide a list of default negative tags to filter out things that might lower the quality of your image. You are welcome to change or add other undesired content to the list.

Sampling Steps

Steps define the number of iterations the AI should refine from the initial creation of your prompt. Higher steps will cost more credits.

Sampling Method

Sampling method is the technique used to transform points in the model's latent space into visual outputs, influencing the quality and diversity of the generated images.

CFG Scale

The scale value decides how closely the AI follows your prompts. Also, when the scale is low, AI tends to produce softer, painterly pictures. We recommend keeping the scale below 7.

Seed

Seed is the numerical value used to initialize the random number generator, ensuring that the same input parameters produce consistent and repeatable outputs across different generations. Copy and paste the current task's seed in the box if you want to create similar images.

VAE method

VAE models can help you adjust the saturations and coloring for your image.

Updated on: 14/03/2025