What’s new
We’ve updated our AI model: Kenna2.0 is now available. Choose it at the top of the Generate Images tool within Workspace, right above Positive Prompt.
How to write a good prompt
Kenna AI can understand natural language, but word order matters most. We can’t stress this enough. Move the most important words to the front, and move words like location and minor details to the end of the prompt.
Try it Out
1. Login to Kenna
2. Click Workspace on the top right
3. Click Generate on the left tool bar
4. Choose the positive prompt area on the right
5. Paste “a red winter coat in a department store” and click Generate
Now try the same prompt again in a different order
“A department store that has a beautiful store display with a red winter coat”
The shift in focus in our language will make a big change in the image. Next, try starting the prompt with a camera angle or art style, or even an artist.
“A cinematic 3D rendering of a red fur coat in a cyberpunk store display”
How to write a negative prompt
Think of magnets: A negative pulls the elements out of an image that you don’t want. For example, if we put a positive like “summer dress” and add a negative like “snow, winter, fur coat” it further reinforces the “summer” of the positive. You can use this push and pull of positive and negative to get exactly what you want.
Try it Out
1. Create an image of luxury purses, and wait for the images to render
2. Observe elements that you’d like to remove, maybe gold buckles or logos. Describe the elements that you want to remove in the negative field, and click generate and compare the images.
How to use weights
Try it Out
1. Create an image with the prompt “luxurious leather wallet”
2. Now try that again but this time let’s exaggerate:
positive: (((luxurious))) leather wallet
negative: [[[modest, affordable, attainable, poor, reasonable, cheap]]]
Use weights when stubborn parts of an image won’t budge, but too many weights can introduce image artifacts. Only use weights when word order isn’t effective.
Limitations
Limited knowledge before 2021
Kenna AI was trained in part with billions of images from the public internet before 2021. As such, some pop culture references may not yet be understood. Please discuss and request any blind spots in our Discord community.
The Generate tool is different from a chat bot
Unlike a tool like Chat GPT, when we are using the Generate Tool, we are not “chatting” with Kenna AI. Every image is a fresh start, so write the full image description every time.
The Generate Tool does not understand prompts like “make a change” or “what if it was like this” commands, , because it isn’t talking at all. Generate is only creating what is in the prompt field.
When it’s time to make an edit, rephrase the original prompt or for spot changes try the Variations and Generative Fill tools instead.