draw command
This page is under construction..
The primary command for creating AI art. You'll use this for doing text-to-image and image-to-image commands. Here are each of the options.
This is where you'll start. Input the things you want to see in the output image.
Input things here to try to remove them from the output image.
Allows selection from any configured models.
The amount of iterations AIYA will go through before completing an image.
How wide the image will be.
How tall the image will be.
This is how closely AIYA will pay attention to your prompt. Higher = pays more attention. Lower = more freedom to let the model take over.
You can select from one of the sampling methods.
This value serves as a basis the generated image. Using the seed can help reproduce the same image or similar images.
This will list the preset styles that have been configured. Styles are like packages of prompts, shortcuts to save on typing. To see what a style contains, click Styles list after using /info command.
Tries to fix faces in images using one of the selected available facefix AI models.
When creating high-resolution images, sometimes undesired results appear (like duplicated anatomy). This tries to fix that by generating the image at a lower resolution (less prone to issues), then upscaling it. The options given are various upscaler models.
Some models perform better with different values of CLIP skip.
This can apply a hypernetwork or a LoRA model that can influence the generated image.
The model's multiplier default is 0.85. To set a multiplier manually, select your desired extra net, then afterward add :# to the field where # is your desired multiplier. The multiplier is a value from 0.0 to 1.0, which determines how much effect the networks have on the model, where 1.0 is the maximum effect.
For example, I have this extra net:
To make the multiplier 1, I'll go back in and edit the field like so:
AIYA will say no options match my search, but it's okay. The extra net and multiplier gets reformatted on the backend. If any bad values are entered, such as letters, it will be set to default of 0.85.
To find out what extra nets are available, run the /info command and check out Hypernets and LoRAs; these are combined into extra nets.
Used for init_img/init_url. Higher value = AIYA is allowed more freedom to alter the source image.
This will take an image attachment as a starter point to generate an image. You can drag 'n' drop or upload an image file via this option.
Same as above, but instead this takes a URL address containing an image. This option will override Init_img, if you're using both for some reason.
Batch allows generating multiple images at once using the same parameters. This single command controls both the batch count and the batch size. Do note that the host may impose limitations on the batch count/batch size for each Discord channel.
- Batch count - Generates images one by one. This is slower, but usually more reliable.
- Batch size - Generates a group of images all at once. This is faster.
There are two ways to enter your batch value: manual, and automatic
-
Manual - To enter your batch manually, you input the amount for both batch size and batch count in the field; the batch format is
count,size(you may also use a.as a separator). Example:
This will place in queue two groups of four images, for a total of eight images. -
Automatic - This is a simpler way to set the batch and is generally recommended. For automatic, you only need to input one number: your desired number of images. AIYA will then try her best to generate that many pictures using the highest batch size. Example:
In this example, the channel has a batch count limit of6and a batch size limit of3. Let's see how she divides my request:
Wowee
The algorithm is not perfect, however. Now what if I ask for7images:
The result is less than ideal (though with this example's limits there is no way to make a batch of7, anyway).
When doing your batch of images, you'll get them in a pre-formatted grid. The reason for this is to avoid how Discord cuts off parts of the image when it does grids. The row of buttons are limited in their scope in that they only know about the overall generation request, not any specific image in the output. To interact with a specific image there are dropdown menus:
This will allow you to grab or upscale an image or set of images from the grid. To use the dropdown menu, open it and select each image number you want. Then when you're done, close the menu and AIYA will perform the request.
For grabbing batch images after the fact, please refer to the /info command.