Showing posts with label AI art Generator. Show all posts
Showing posts with label AI art Generator. Show all posts

2026-02-11

Variations of a mermaid

 Quite often using the same prompt with different AI Generators will result in getting images which are very different from another. 

This image rendering a cute mermaid was created with Nightcafe Studio. Quite a happy-go-lucky mermaid, with pointed leaves as a kind of hairdress.









The second image was created with Leonardo AI. Take a look at the leaves coming out of her back! And the long strands of hair which were added by the AI Generator. 

By the way, this was the only image out of four which were created - due to moral restrictions.







The last image, created with Google Gemini, shows not only a different pose but also a different headdress. 

2025-08-30

Two version with one prompt

Sometimes I like to work with several AI Art Generators and one prompt, and everytime I am surprised how different the results can be. 

The image on the right was created using Piclumen AI, which has a nice free version. You are getting reward points by uploading your images to social media (pinterest for example) and liking and commenting other people's creations.


The image on the left was created using Nightcafe Studio. Here you get reward points by showing up every day.


2025-04-04

I just discovered Cogniwerk AI

 

There is a new AI Image creator online almost every day. I usually get myself a free account, experiment with the settings and then decide to either use it on a regular basis or only now and then.

Cogniwerk AI is one of my "newer" AI Image creators. To create one image costs one credit; every account has been given a certain number of credits, and when you have used up your credits, you can request more. 

Many AI Image Creators have difficulties with horses - they tend to create animals with five legs. I wanted to know what Cogniwerk does with a prompt like this:

A black mustang galloping over a yellow green meadow, vector style

Well, it actually added an additional leg where it does not belong. However they offer Inpainting, and so I managed to create the anatomically correct mustang in the image above. So up to now I am very pleased with Cogniwerk AI and will use it regularly.


2025-02-20

Starry AI

Sometime last year - or even the year before - I discovered Starry AI. I wrote down the webadress, but other than that I didn't do anything with it. A few days ago I found the note with the address on it, and tried it out. I must say that I am very pleased with the results I am getting. 

First of all, I have the free version. The free version has got its limits regarding the format: you can only create images in portrait format. But: you are able to get five lumens (or credits or tokens) by logging in each day. You are also able to earn lumens every three days by posting your images on instagram or X (twitter). And, as I mentioned above, the results are very, very nice!

I wanted to create an image showing a cute rat carrying a bowl with coffee beans. I first tried Nightcafe Studio, but didn't quite get what I was going for. So I used the magic prompt Nightcafe had created and applied it on Starry AI.

The image on the left more or less shows what I had in my mind. This is one of four images and the one I liked best. 

So Starry AI has become one of my favorite AI Image Generators, together with Nightcafe Studio, Bing and Mage Space.


2024-02-14

Stablecog

Stablecog is an AI Art Generator which allows the use of Starter images. However each time I tried to do this the result was disappointing. Of course this can be due to my not using Stablecog in the proper way, but there are AI Art Generators which I had been more success with so far. 

But this does not mean that you cannot get nice results by simply using a prompt. Here are two images I created with the prompt: "beautiful wooden carved ornament, blue shades, intricate".






2024-01-31

Cute Lioncubs

 I definitely like cute animals. So I try to generate them with AI Art Generators. Here are some examples:

Here's a kitten holding a red heart, created with Full Journey. 

Full Journey once was Bluewillow AI. You get 4 images with one prompt, and you can use the images commercially. Full Journey is not totally free - you are paying for each prompt with one credit. However your credits are filled up to 50 every month or so. The quality is quite nice, and you can choose the aspect ratio. 



This cute lioncub was created with Nightcafe Studio. Nightcafe Studio is not free either. You pay 1 credit for one image. However, you can earn additional credits: You get 5 credits for logging in, 3 credits for publishing your image on instagram, 2 credits for voting in the daily challenge. In addition to that you get 5 credits for every five images you create.

The creations can be used for commercial purposes (for example as designs on print on demand websites).



The third example of a cute lioncub was generated with Bing Image Creator. Bing only offers 1:1 aspect ratio. At the beginning you could read on the website that commercial use is not allowed; in the meantime it looks like there is no such explicit restriction anymore. 

Bing is free for anybody who has a Microsoft account. You do not have to pay any credits; you have so called booster points which help to generate your images faster. You can also earn additional booster points by answering certain questions.






2024-01-22

AI version of one of my latest paintings

In a recent blogpost I told you about a painting I did with my PITT Artist Brush pens. Of course I wanted to create AI versions of it. 

One of the AI Generators which allow the use of start images is Limewire AI. In 2023 it was called "Blue Willow AI", and it worked only on discord. 
At the end of 2023 it changed to Limewire AI and you can now use it on a web interface.. 
Blue Willow was totally free; now on Limewire you can create 10 images daily. 
I love how this image came out. It still has the original colour combination, but the word "cubism" in the prompt did a nice job.



2024-01-17

How to make an image semi-transparent with Canva

 Sometimes you need to make an image semi transparent in order to use it for print on demand purposes. 

I created this red hearts design with playground AI. I wanted to use the image for designing a sticker with a text on it on Zazzle. As you can see the readablity of the text would not be very good due to the opacity of the hearts. Now Zazzle does not have a tool with which you can change the opacity / transparency of an image. 

So what could I do? Well thanks to YouTube and Mey Aroyo's channel I found the solution and of course had to try it out myself on the spot:

Using Canva solves the problem. You create a file the size of your image, upload the image, put it on the file. You then select the image. On the top of the image you see the symbol for transparency. Clicking on it opens a slider with which you can change the transparency to your liking. You can see the result in the image below. You can then download the image, enlarge it, upload it to your print on demand shop and put a text on it. Have fun!



2023-12-01

Promptalot

Promptalot is a website for organzing and sharing prompts for Midjourney AI. I don't use Midjourney, but many other AI Art Generators. I am a very curious person, and so I want to know what happens when I use one of these prompts with Nightcafe Studio, for example.

The first step is to go to Promptalot and search for a subject. I searched for "Pattern" and one of the prompts I got was:

A digital artwork depicting a complex fractal pattern with a symmetrical design. The fractal features an ornate center resembling a blooming flower with a warm, luminous core of yellows and reds. This central motif is surrounded by intricate petal-like structures in a palette of blues, purples, and greens, suggesting layers of depth and texture. The background simulates a misty forest at twilight, with the soft silhouette of trees and a hint of a sunset in the distance, contributing to the mystical atmosphere of the composition

I copied the prompt and went to Nightcafe Studio. The result:

It definitely does not look like the image created with Midjourney. Nevertheless I like it. 

Now the question is: How about using Leonardo AI, or Ideogram AI, or Dreamlike AI or Deep Dream Generator AI? 

Or how about changing the colors in the prompt? Or shortening the prompt? 

For me the goal is not to create the same image as Midjourney would. For me the goal is to use the prompt as an inspiration to create something different, to play around.

2023-11-08

Ideogram AI

 When I look back one year, there were not many AI Art Generators in November 2022. I learnt about Midjourney, but it was not free, and I did not like the fact that it runs on Discord. I learnt about Nightcafe Studio and I immediately loved it and got addicted to the whole field of generating images with this Art Generator. 

Now, one year later, the number of AI Art Generators is still growing. A few days ago I watched a video in which Ideogram AI was given a try, and of course I had to try it, too.

I used the following prompt:
black cat in a colorful garden, folk art style, intricate

And here is the result:

You get four images with one prompt. The images can be saved on your profile page. Ideogram is free. You are offered a variety of styles,

I used this image as "a parent" and added a style to it.

Now the prompt is:

black cat in a colorful garden, folk art style, intricate, painting.

I can only invite you to visit Ideogram and play with it!

2023-10-11

New AI Generator: Full Journey

There are lots of AI Image Generators on the web, and there are a lot of changes going on. I once used Lexica, which now seems to be almost dead. I used Clipdrop for quite some time, and since last weekend their website seems to undergo a lot of changes and /or bugs.

On the other hand, there is always something new to try out and explore. For example there is Full Journey AI. It is a bit similar to Midjourney because it only runs on discord, but the commands are different. There is a free plan which gives you 50 tokens / credits every month. 

I quite like the quality of the images. You get four images with each prompt, like with Midjourney AI.

The prompt for this image was:

still life with crystal glasses, in the style of Rubens -43

The second thing which I was very pleased with was the quality of the service; I asked a question regarding how I could get the info about my remaining tokens, and it did not take more than 5 minutes untill I got the answer. 

2023-10-04

Uncrop tool in Clipdrop

If you are using Bing Image Creator you will have noticed that it only offers the square format for the rendered images. The quality of the images is quite good. However I like to put my photo prints into frames, and the standard frame for photos are 4:3 or 3:4 sizes. 

Lately I watched a video about Clipdrop and I learnt that Clipdrop has a tool which makes it possible to change the format on a rectangle image. It is called the "Uncrop" Tool. I had to try it and was very pleased with the result. 

This image was created with Bing's Image Creator. I love the colors, and the flowing shapes. I'd like to use it for a selfmade card or frame it and hang it on my wall. 

But, as I said above, I could not find any frames for the square format.

So I had a go with this image and the Uncrop Tool on Clipdrop's website.









This is the 4:3 version. Yes, the image is smaller than the square one: I have a free account and that means that the image is downsized when using the tool. 

However, Clipdrop itself offers a tool with which you can enlarge your image, so that will be no problem. 


2023-09-22

Blue Willow - a place to learn how to prompt

 Blue Willow AI is only running on discord. For some people this might be a disadvantage, because not everyone is familiar with the way Discord is running. However, now that I've made myself familiar with it I quite like it. One of the reasons why I like it is that you can easily learn from other people. You see their images and you can read their prompts. You can easily copy their prompts and change them. And you can easily change your own prompts and see what effect the changes have. 

The prompt for this image was:
Realistic, 8k, dolphin jumping from water, --ar 3:2

There is a bit of a sunset, but I wanted to have more of the sunset in the image. So I copied the text and added "sunset" to it.

You can see the result in the image below. 

By the way: I do not recommend only copying prompts from other people; the fun - and the learning - begins when you add things to the prompt, or replace words, with other words: play with the prompt! 


2023-09-04

Promptcrafting: The order of words is important!

 

The prompt for this image on Openart AI was:

painting of a stack of pumpkins in front of a cottage in the style of Anne Stokes

Well, this image looks like a hoard of pumpkins is going to overtake the nice little cottage house. It could be used for a horror story like "The night of the killing pumpkins" or "Pumpkins took revenge for being mistreated by human beings". Not at all what I intended to create...

Now, one has to keep in mind that the word order is very important to AI Image Generators. 




So I decided to change the prompt into:

beautiful cottage house with pumpkins in the garden in the style of Ivan Bilibin, Ernst Haeckel, Daniel Merriam, watercolor and ink

The words "beautiful cottage house" in the beginning of the prompt leads to the fact that the house is actually in the focus and you can see that the pumpkins have about the right size.

2023-08-30

Vectorize AI

 In some cases Vectorizer AI is the right tool for enlarging images created with AI Image Creators. 

Here's an example:


I generated this troll with the use of Dreamlike AI. The dimensions of the image are: 576x704 pix, much too small for any use on print on demand stores.

I had learnt about Vectorizer AI a few days ago and also watched a video. I am always open for anything new, so I wanted to try it out.

Now step 1 is to have the background removed, together with the strange parts on the right. There are some background removing tools online, I went for clipdrop's tool.

Now step 2: upload the image to vectorizer AI. Here the result:


Now the size is 2304 x 2816 px. You probably won't notice the changes on the image much, but you will see them when you open it with a picture processing program like Photoshop or Gimp.

Vectorizer AI is free and it is still in Beta. It does not work with every image; but I will try it again!

2023-06-30

How to deal with the big number of images in PicFinder

 Most of the AI Art Generators I use create four images with one prompt, and most of the time I can use one or two images out of the four. Most of the AI Art Generators also save your images with your profile - so you can come back and look at them and decide which one you want to download. 

PicFinder is different: It generators far more images than any other art generator, but it does not save them with your profile. So you have to download them in order to save them for later usage. You have two options: you can either download them as 512x512 px files, or you can enhance them and download the enhanced version (which takes more time of course). 

It took me a bit of time till I found a satisfying solution for me. I quickly go through the images and download all those I find interesting. 


For example, take these three Indian elephant images:On first glance I like all three of them. However when it comes to make prints out of them I'd rather prefer not the front view (the tusks are much too yellow), but the other two. But I am not sure yet. 

So I keep them in the original size. I can always go on:

https://www.pixelcut.ai/image-upscaler

and upscale / enhance them.


By the way: From the beginning of my AI Art activities on (which means from November 2022) I use the name of the art generator or an abbreviation in the title of the image file. You will ask "why?"

Well, some art generators allow commercial use and some do not. In order to be on the safe side I now can see where the image file comes from and can use it commercially or not. 

So in case I plan to use an image file as design on my Zazzle shop (for example to create a puzzle with it) I know exactly which images I can do this with. 

2023-06-28

Stablecoq AI

 In the meantime the number of AI Art generators which I am using to create images with has been risen to nine:

  • Nightcafe Studio
  • Lexica AI
  • Blue Willow
  • PicFinder AI
  • Bing Image Creator
  • Stablecoq AI
  • Color Pop AI
  • Deep Dream Generator
  • Dreamlike
All of them I use without paying anything. There are some I use every day, and there are some which I use not so often. 
Stablecoq has become one of my favorite AI Art Generators. 
1. Quality of Images
The image on the left - a portrait of the pharao Ramses - is a good example of the excellent quality of the images Stablecoq is generating. 

2. Various Aspect Ratios
It not only offers Square, Portrait and Landscape formats, it also offers aspect ratios for Mobile and desktop.

3. Various Models
I must admit that I haven't played around with the models yet - always work with the standard model displayed which is Kandinsky.

4. Search for Images and Prompts
This is most helpful - not only when looking for a certain prompt but also as an inspiration for generating AI Images in general.

All in all I can only recommend this AI Art Generator!

2023-06-23

One prompt, two very different images

 I have been using quite a number of Artificial Art Generators. In the last weeks I've added Stablecoq AI and PicFinder AI to my toolbox. Both are free to use. Stablecoq AI works with tokens. I found out however, that every time I use it tokens are added... This is a bit strange, but I take it as it is. 

Now with PicFinder you get numbers and numbers of variations of an image. These images are not saved to your profile - so you have to look through them and decide which ones you want to keep. You can then download them in the original size or download them in an enhanced size. Once you leave the page the images are gone. 

I now tried the following prompt with both Stablecoq and Picfinder:

beautiful anatomically correct fairy princess, anatomically correct no hands, Dappled Fashionable Fairy tale, Idyllic by artist "Carlo Crivelli", 3D shading, elegant, ethereal, fantasy

Now you might ask "why 'no hands'?" Well, AI Generators still have problems with the rendering of hands; so I decided to ask the AI to go without.

Here's the image created with Stablecoq.


As you can see, no hands, but otherwise quite the result you would expect with the prompt.

Carlo Rivelli by the way was an Italien Renaissance painter who lived from 1430 to 1495.







And here is the image created by PicFinder (well, one of the many images; I must admit I had difficulties to select one image.


I was very much surprised that the fairy on the image had anatomically correct hands with the correct number of fingers. 

Up to now I have not looked up paintings by Crivelli in order to find out which image is more similar to his style; it is up to you to do this. But in anycase I am very pleased with the outcome.

I think these two images are a perfect example of the various ways the different AI Generators interpret prompts. I can only encourage anyone who is interested in creating AI Art to try out the same prompt a few times with other Generators!

2023-06-16

Color Pop AI

 Since the time I have been generated my first AI Images I have been looking for a way to generate coloring pages. I know that many people are selling their coloring pages, but for now that is not my intention. Coloring pages are a wonderful way to try out color combinations with alcohol based markers or practise the way of applying paint to the paper. You will find a lot of free coloring papers on the internet.

However I want to use my own designs. Today I learnt in one of the facebook groups I am a member of about a new AI Generator which generates coloring pages: Color Pop AI. Of course I had to try it. I submitted the prompt: "cute little elephant", and in a very short time, four pdfs in the size of DIN-A4 (letter size) were created. These pdfs are not saved with your profile, so you have to download all the pages you like for later use.


After you have downloaded the pdf you can use free online converters in case you need the pdf as a jpg or png. 

Or you can transfer the drawing with a light table to a thicker surface, for example watercolor paper and have a go at it with watermedia.

2023-06-14

Picfinder AI

 The number of AI Art Generators is constantly rising. In one of my facebook groups the moderator mentioned Picfinder AI and recommended it. I was curious, went there and opened an account. With an account you get 1000 images daily. 

I inserted "tiger in the jungle" as a prompt and got a huge number of results. I selected two of them:


You can choose whether you want a square format, or portrait or landscape. The original square is 512 x 512 pixel; it says that you an download an enhanced image - but this didn't work when I tried it. So I downloaded the original size - which is okay for demonstration purposes in this blogpost. In case you need a bigger image size there are many AI image enhancers on the internet - one of my preferred ones is pixelcut/image-enhancer which is free.




As you can see, the anatomy of the tiger is correct - no additional limbs, ears or other anomalies.

I certainly will play around with the new AI Art tool!