Here we briefly explain how to create and edit images in the chatbot @chat_ai_tg_bot. These instructions also apply for working with images on the website.
To create an image by description in the chatbot, send a message “/image description” or “draw description”. Example: “draw a cat and dog wedding”. Works in both Russian and English.
Model selection
To create an image on the site, go to the chat 🖼️ Create Image and send a message which you can immediately start with “description”. Example: “cat and dog wedding”. Works in both Russian and English.
Create image
After submitting the description, you can choose a model. We will look at the most advanced model, Journey.
After choosing the model, the download starts and after a short while, you receive 4 ready images with buttons.
Images by description
By pressing on 🔄 - the bot will send another 4 photographs by the same request.
By pressing the V1, V2, V3, or V4 button - 4 more similar images will be created depending on the chosen number: 1 - is the first photo on the left, 2 - is the right one at the top, 3 is right in the middle, and 4 is right at the bottom.
Result of pressing V1
By pressing the U1, U2, U3, or U4 button, the bot will send a specific photo and additional actions.
Result of pressing U1
Here you can get new similar images by clicking “Change significantly 🪄” or “Change slightly 🪄“. You can also zoom the image on all sides by clicking on “Zoom out 🔍” or only on one specific side by clicking on “⬆️⬇️⬅️➡️”.
Result of clicking Zoom out 🔍
You can edit or create similar images using your existing photos or graphic works.
To do this, simply send an image with a description to the bot or to the site.
Sending image with description
To work with Journey click "Edit 🪄"
Result request - Neznaika in anime style
Further, you can also edit and enhance the obtained images using buttons.
This feature allows you to use any picture as a base style. That is, you can write a request, upload an image, and the output will be a generation that considers the stylistics of the uploaded image.
Copy Style Journey 6
Request: Ronaldo football player
You can upload your photos and any other face you want to use.
Send your face that you want to use to the bot. And choose Face swap
Face swap
Afterwards, you need to upload a photo where your face will be placed.
Second photo
A few seconds — and your photo magic is ready.
Result
Or upload to the site the face you want to use. And choose Face swap
Face swap
You will have the opportunity to upload the second photo.
Second photo
And here is the result.
Result
Use the request auto-improvement button in the bot.
Enable request auto-improvement
Or the Improve the request button on the website.
Improve the request
Our algorithm will add a more detailed description, styles, and the needed format for your request. You will receive much more high-quality photos.
Fisherman on a boat
The bot works best with simple, short sentences describing what you want to see. Avoid long lists of requests.
Incorrect: “Show me a painting of many blooming poppies, make them bright, saturated oranges, and draw them in an illustrated style with colored pencils”
Correct: Bright orange poppies, drawn with colored pencils
It’s better to describe what you want rather than what you don’t want. If you ask to organize a party “without a cake”, your image will likely include a cake. If you desire to ensure that an item is not present in the final image, try using the advanced prompt with the --no
option.
Incorrect: “Party without cake”
Correct: Party --no cake
If you want to get an image of a specific size, you can specify it in the description through the command --aspect
.
Example:
Party --aspect 4:3
Party --aspect 16:9
Party --aspect 1920:1280
If you have a link to an image, you can also add this link to the description.
Example:
“/image link to the picture “anime style”
Creating images is available by subscription.
You can purchase a subscription by sending the command /subscribe.