Midjourney AI Image Generator: Beginner to Advanced Techniques and Parameters

1. Course Introduction - What you'll Learn: This course, I will guide you through the fundamentals of me journey. You will learn many different techniques for generating AI images using one of the most advanced AI tools. We will also look at the commands that you can use, the different compositions, settings, textures, models, and many more features. We will also look at the main parameters, and this is where you can customize the type of images that you want to be generated. For example, the aspect ratio, image quality, image weight, and many other customization features. We will also look at different prompt builders that can help us generate specific types of images. This course also has a project section where you can share your own AI generated images and include the prompt that you used to generate them. 2. Midjourney Guide: This section, we will talk about how to use mid Journey and many of the different functions. But before I recommend going over to dogs journey.com where you can look at the mid journey documentation. So for this, we can click on this one, the Quick start guide. This guide is basically looking at the instruction manual of med journey on how to get started, how to use this chord, what are some of the commands, parameters, tools, how to write the prompts, how to look into the styles, the aesthetics, and also your subscription. Before starting, I recommend looking into the instruction manual and keeping it handy because med journey has a lot of different features that in some cases, you might need to refer back to the instruction manual if you want to memorize them and if you want to know how to use them. 3. Midjourney Intro: So now we're going to start using MD Journey. For this, we're going to go over to Google and we're going to search for three things. The first one is going to be Mid Journey. This one is going to open up the original website. From here, you can always sign in and you can sign up for a paid subscription. On this one, we can go over to the signing button and we're going to sign up with this Cord and we're going to authorize it. Now we can purchase a med journey subscription, and there's a free version of med journey that I have never been able to use. It just never works because there's always too many people that want to use it. I basically recommend going for the paid subscription. Otherwise, you might actually not be able to ever use it. For this one, I'm going to go for the monthly billing and I'm just going to choose the standard plan. There you go, I have now subscribed to the paid subscription. So this one basically gives me 15 hours of fast generations on limited relaxed generations, general commercial terms, and a few other things. So here I have the mid Journey butt. But if you don't see it, you can go over to Google once again and you can search for the Mid Journey server and click on this one, Md Journey. This one basically invites you to join the Mid Journey server. And now you can just accept the invite and there you go on this side, you will have access to the mid Journey server. And here you have some options. For example, if you go over to the newcomer rooms, you can see what other people are generating, and you can also generate some pictures right here. You can even see the prompt that they wrote and here you can create your own images. However, they will be shown to everybody in the server, not just you. Something that you can do is if you only want them for yourself, is that you can create a new server, and I'm going to select create my own for me and my friends, and I'm just going to call it the med Journey private I'm just going to create it. Now I basically have a normal server just for myself. Now we have to invite the med Journey bot. We can either do this two ways. We go over to the med Journey server, we click on the Mid Journey bot now we can just click on add app, then we can select the server. In this case, I'm going to select med Journey private, and I'm going to continue. Then I'm going to authorize it. Then I'm going to say that I'm a human Now I can go over to my server. Now if you look over on the member list, we have myself and the Mid Journey bot. This is one way to do it. The other way is to just go over to Google and search for the mid Journey bot and select this one, the one from this Cord, you can click on it, and here you get the information of the bot, and you can click on Add to server and then just select your server. In this case, mid Journey private, and then continue authorize it. And it has been added to the server. You can either do it one of these two ways. If you cannot find it on the mid Journey server, you can look for it on Google and you can find it. That's how you get access to Mid Journey. 4. Basic Prompts (/imagine): Now we're going to look into some basic prompts for mid journey. If we go over to the documentation, here we have a section for prompts Prompts are basically just like the prompts that you give CG PT. But in this case, you don't have to make them very long. You can shorten them to just exactly what you want. We're going to look into basic prompts. For this, we're going to write the backslash imagine, then here we have the text prompt section where we can write what we want. You can also look into advanced prompts where you can even put your own image, text and the parameters, which we're going to look at a later lesson. On this lesson, we're going to focus on the basic prompt function, which is basically just putting some text and generating images from that. Let's go over to this chord and here you have two options. You can either use the private server that you created, you can create a new one as well, or you can go over to the mid journey server, and you can select one of these newcomer rooms, but keep in mind that you're going to be sharing them with random people. So if you click over here to the member list, you can see who is going to be able to see your prompts, and you can also see their prompts. In this case, I'm just going to go over to the private server that I created, where it's just me and the mid journey but and I can start creating here. If we go back to that documentation, we can see some things that you can think about. For example, you can think about what subject do you want, what kind of medium, what environment do you want? What lightning, color, mood. The first one I'm going to try is just a simple subject. So the way you create an image, and this is the most important part is that we write the slash key, and we can see some of the med journey settings. The one we're going to use is going to be the imagine one, which basically allows us to create images. If you don't know what they do, you can look at the description here. The first one is to answer questions. The third one is to adjust your settings. We want to use imagine, which is the one to create images with med journey. And it's basically going to create this section for us to write our prompt, and I'm going to type Osaka Japan in a dream like scene. I send it, and here we can see that our image is being generated. Here we go. We now got four images of Osaka Japan in a dream like scene type. I think they look quite nice. But now I want to try a different prompt. If I go over to that documentation, we can see that we can also change, for example, the style, like the medium. Do we want it to be a photo or a painting, a sculpture? I want to try a new prompt and I'm going to write imagine. I'm going to write an ko style, so realist painting. This is the medium that I want. Then I have the previous prompt of Osaka Japan in a dream like scene. Now I send it. Now we're going to change the medium that it's done. And there we go. We now have it in ko style, which is a painting style. We have our normal prompt that is just very basic. Then we changed the medium and now it looks like this. Now I want to try a different format. If I go back to the documentation, we can also see that we can change some of the environment. Now I want to add some more details into this one. I have my previous prompt of the surrealist painting of the pen. I want to add at the end that I want it with flying islands. On this case, I'm adding more things into the picture. So it's going to be similar to this one, but with flying islands. Let's see what we get. There we go. We now have our pictures. I think I can see some flying islands on the background of our image. So this way, you can always add more detail, add more things into your picture and met journey will generate it for you. The last prompt that I want to write is this one, imagine. I'm going to write the same prompt, but I wanted a bird's eye view. I also want it in a calm mood. I'm adding two more factors. I want a bird's eye view and a calm mood. I think it's not going to be very different from this one. But let's send it. There we go, we got new images, and this ones look pretty nice as well. You can see that we can keep on adding more details and details and met journey will generate whatever we tell it to. That's how you create some basic proms with mid journey. 5. Upscaling: We're going to upscale our image in mid journey. If we look at our images that we have created in mid journey, we can see that here we have some options. The letter U means that we can upscale those images. The way images work is number one, is this one, top left, top right is number two. Bottom left is number three and bottom right is number four. If you click on the one with the letter U is basically going to upscale that image. If you click on the letter V is going to make variations of that image. If you click on this one, is going to regenerate them. Let's try to regenerate them. Let's see what happens. So you can see now we got more images, but it's still using the exact same prompt that we gave. When you regenerate, it will do something different, but it will use the same prompt that you gave it to. These images look a little similar, but they're also different. They're still keeping the same similarities that come from the prompt. So now let's try to upscale them, and I'm going to say that I like this one, number two. I click on U two, and it's not going to upscale this image. There we go. We now have a bigger version of this image. If we click on it, we get this image. However, I do not recommend to right click it and save the image right here because it's going to save it as a web image. The best way of doing this is to click on this one, opening browser, Now you can download the image. Now you can click on Save Image S, and then you can save it. Now you can save it as a dot PNG. If you were to save it straight from this chord and you click on Save Image S and you save it, You're basically going to get it as a web P. This one is not the best format. The best format, I think is PNG format. Now we can either choose to upscale it even bigger if we select times two, it's going to take twice that time, but it's going to make it even bigger or scale times four, which is going to make it even bigger. Let's try to upscale times two. This one took a little bit of time to come up, but now we have a much bigger image. With much more detail. If I open in browser, the image looks way bigger and it has way more detail. I'm just going to save the images and save it. If I go, I can see that this image, the newly created image is 7.7 megabytes, which has much more detail. The we p version only has 1.7 megabytes, and the normal one, which is the other P and G one, only has 2.1 megabytes. There's a noticeable difference between these ones. This one is the previous one and the newest one, you can see that it has way more quality into them. If you really like an image and you want to have it with much more quality, you can upscale that image, you can also upscale it to four times and make it which have much more details. You can either upscale two times or four times or just leave it like that like normal. 6. Variations: Now we're going to create variations of our image in mid journey. If we go back to the previous images that we have right here, I selected number two to upscale it, but now I want to make a new variation of this image. First, I have to select what image I want to make variations of. Once again, image number one is this one, number two, number three and number four. I'm going to make variations of image number four. So for this, I can just click on B four, and it will start to create variations of that image. There we go. We now have four images that are somewhat similar to that image, but there's some variation into them. If you want, you can keep on clicking B one, b22 or four, selecting your image to make variations of that kind of image. Something else that you can do, for example, I'm going to select to upscale the third image right here. I click on three and it's now going to upscale it Now I want to make variations of this image, so I can either do a strong subtle or region. Let's try the first one, very by region. Now he's going to ask us what we want it to be different. I'm going to say delete this region right here, I'm going to submit it, and let's see what we get. You can see now we can see our image, and now it's generating this section. At the moment, you cannot see it because it's at 31%, but it's going to start generating it. And there we go. We have four different variations. You can see that the left side is still there, but the right side has changed and we have four different variations where the right side is different. That's something that you can do. If you click on the vary by region function. Let's try this subtle function. There we go. We got four very subtle variations from this image. I cannot even tell the difference, but you can see that, for example, the temple is a little different and the bridge is different. The background is a little different. But they're not so noticeably different, they're very subtle variations from our image. Now I'm going to select this one where we get strong variations. There we go. We can see that the place is still the same, but the image has way more variations into them. This one, they all look very similar. The variations are very subtle, but this one, it made more changes, so you can see that there are bigger variations right here. This is how you can use variations to make some modifications to your images. 7. Zooming Out: Now we're going to use the summing out function. Let's go over to mid journey, and I'm going to write slash imagine, and the prompt is going to be a superhero action scene with characters fighting and using their powers in a digital comic book style with bold colors and dramatic angles. Here I have a big prompt. I send it and here we get our results. From this one, I like, they're all pretty good to be honest, either number one or number two. Let's go for number one. Now I'm going to upscale number one. Here we go. Now we can use the soming functions to make it look better. We can either som by two times by 1.5 or a custom Zoom. Let's start by clicking on Sum out times two. I click on it. Here we go. It's now generating more variations and it's now more summed out. This one is the original one. Now this is the one with the Som function. I think that looks pretty good. Now I'm going to select the number one again. I like this one. And if you want, we can always Zoom again, for example, two times or 1.5, or we can select a custom zoom. If we select custom Zoom, we can modify some of the parameters. Here we have the Zoom parameter, and we can change for example, instead of two, we can select one, and we can submit it. Let's see what we get. There we go. We now got an extra zoom by one more point. So you have the same image, but it's sum back a little bit more this time. This is how you use the oming function, you can add more details and you can keep on expanding your image and summing in backwards. However, one thing to notice is that whenever you use the custom Zoom function, the maximum is two. For example, if I select three and I submit it, I'm going to get an error because the value of the om has to be 1-2. Keep that in mind whenever you're trying to som out of your images. 8. Panning: Now we're going to use the panning function. I'm going to use the previous image that I created before and that I also summed out. I'm going to select a variation. In this case, I'm going to go for number one again. There we go. We now have our image. What I want to do now is that I want to panic and here we have the functions. If I click on the left keys going to add more things to the left of the image. If I select the right key, is going to add more stuff to the right side. If I select the top, it will add stuff to the top and bottom, it will add stuff to the bottom. On this image, I want to add stuff to both the left and the right. I'm going to click on left. You can see right here that it starts to generate something to the left of the image. The right side is still normal and the left is adding something to it, and this looks pretty random right here. There we go. Let's see which one I like the most, I think the most natural one is probably going to be the third one. But let's go for the fourth one. Let's see what else we can do. I'm going to click on upscale number four. There we go. We get a few more people right here, more characters fighting. Now we can only keep on panning either to the left or to the right. I'm going to add something to the right. I click on the right side. Now you can see that it starts to create something on the right side. The normal part is left untouched and it's just going to add a few more things. Check it out. Now our images are wider, and we get a whole bunch of new characters coming up now. From this ones, I would say that I like this one, the second one. You can see how we went from a simple image that is this one with a few characters, and we went over to one like this with way more characters. From this ones, I like the second one, so I'm going to upscalt there we go. This is the new image that I generated. You can see that we started with a simple image, which is this one, and we ended up with an image that has way more details, way more characters. Which is this one. If you want, you can also try the top and bottom functions. We go back to our original image that allows us to add up panning and down panning. Let's see what we get. Let's go up. This is what we got now. We've got a few more details. We now get some superheroes on the top. And our superhero down here. This is the new image that you generated. If I select any of these ones, for example, number four. Let's up scale it. There we go. Now we can keep on modifying it, but we can only pen up and down. We can no longer pan to the sides. It basically depends on what you select first. If you select the sites first, you can only keep on expanding to the sites. If you select up or down first, you can keep on panning up and down after. But of course, you can always keep on summing out. Let's try the sum out 1.5. Now you can see that it's going to generate stuff on the top bottom and the two sides. There we go. We now got our images. Now it added more details to every side, top bottom left and right. If I choose to upscale an image like image number four. Now it allows us once again to be able to move to the left to the right, up and down. These functions are back. This is how panning your image works. 9. Image Prompts: We're going to add our own image into med journey and add prompts to that image. For this, we can go over to the med journey documents, and you can read some of the information about image prompts. Here we have our prompt, and the first part is going to be the image prompt, then the text prompt that you can write the last part can be some extra parameters that you can write to your prompt. The way it works is that first you have to upload your image to this code. Here I'm on a server, which is the mid journey private server that I created. If I want to upload a file, I go down here, I click upload a file, and I'm going to select my profile picture, and I'm going to send it, then I send it. Now we have our picture. I'm going to click on the slash key, then imagine. The first thing I have to do now is to drag the link of my image over to where it says prompt and release and is going to write the whole link of the image. Now the next part, as you can see, we now have the image prompt section. Now we have to do the text prompt and the parameters if you want. Now let's write a text, and I'm going to write a 30 year old man walking down the street in Tokyo retro anime style. I send it There we go. We now have our images, and you can see that they look very similar to our original image. But it's also keeping the information that I gave on the prompt. It basically looks like an anime retrostyle of Japan. Now let's try another one and I'm going to add another image. I'm going to select this one, which is a painting that my mom made. I'm going to select it and send it. There we go, we now send it to the bot. I'm going to write slash imagine and the prompt. The first part is going to be the link right here. Now I can write the text prompt, which is going to be watercolor angel fighting a demon. Very similar to this anyways. Let's send it and see what we get. There we go. You can see that it's keeping some of the similarities to the original image. For example, here we have an angel with a sword and a demon, and this one looks very similar. You have the wings, the sword and the angel. You can see that the new images are very closely related to the original image. They look quite similar. Now let's try one more, and I'm going to upload another image, and I'm going to select something really bad, which is basically like a logo that I made on my iPad. Let's see what we get. There we go. Now I'm going to write this slash, imagine put in our image link. The prompt I want to do is this one, a logo of a bakery. Very simple. Let's send it and there we go. We now got some logos. This is the original logo, which looks really bad. Here we go some logos that are somewhat closely related to the original one. From this ones, you can see which one you like, and T one is the original one and tries to come up with something somewhat similar. If you want to change, for example, like the letters, you can always do that in photoshop. This should be very simple to do. 10. Multiprompt Commands and Weights ( :: ): Now we're going to use multi prompt commands and weights. Here we have the information about multi prompts where it uses a double colon as a separator between two words. For example, you have some words like space ship, and if you write it, you're basically going to get a spaceship. But if you use the separator, you can separate space and ship. For example, this one is the spaceship, and here we have space and a ship. The same thing happens, for example, for cheesecake. If you put cheesecake, you get a cheesecake. But if you separate them, you get cheese and cake. If you separate them, for example, cheese cake, and painting, you get a painting of all of those three things of cheese cake, and paint. Let's try the first one. I'm just going to write a simple cupcake, and here we get some images of a cupcake. But now let's try it the other way. I'm going to write cup double colon and cake. I send it Now we get a very different result. You can see here we have a cup and inside a cup, there is some cake. Basically, a cup and cake, two separate things. Now let's try a different one. I'm going to write cup cake and rock statue and see what we get if we use the three. Let's see what we get, and here we get some new things. This one looks like a rock statue. In some cases, I don't see that I see that cup right here, but I don't see that cake. Maybe this one is a little bit of cake, but in two of the others, I don't see them. In some cases, it might actually fail. Keep that in mind if you're trying to put too many things together. Something else that you can do. Is that you can add some weights to the prompt. Basically, you add a number next to the colon, and that is going to give it more importance. In this case, if you write space ship, you get space and a ship. But if you write space equals two and then ship is going to prioritize space over the ship. You can just put any number actually. In some cases, you might put one and two In other cases, you might try like a number like 100 or cheese 100, cake 50. It basically prioritizes whatever number is higher. Let's try the same prompt right here, cake and Rock statue. In this case, I want to prioritize the cake. I'm going to say cake, I'm going to give it a priority of three, a priority of two. Rock statue, I'm just going to leave it at zero, which is a default of one. I send it. Let's see what we get. It should prioritize cake, then cup The last one should be a rock statue. Yeah, this looks pretty good. I can see definitely weigh more cake in this one and a little bit less of a cup. But it's still a statue, a rock statue. Now let's try another one, and let's try this prompt. Imagine prompt meat and ball. Since Ball has the number two, is going to prioritize ball over meat. Send it. Let's see what we get. In this case, it's another fail because I do not see any meat, but I see some balls. Let's just give it a last prompt. Imagine meat ball a simple meat and ball. Let's see what we get. There we go. This looks very weird. This looks very weird, but I can definitely see some meat and some balls. This one looks horrible, but keep in mind that in some cases, your prompt will work in others, they might not, so you might need to add some weights to it. In other cases, it might just go crazy and look like this last one. 11. View Perspectives: Now we're going to look into different perspectives that we can put into our prompts. What I mean by perspectives is that we can have a prompt and we can either select some perspectives like a close up perspective or an aerial view perspective or a silhouette type of perspective or a dash cam perspective. If you can think of any perspectives, there are two recommendations. The first one is that you go over to Google and you search for different names of perspectives that you can put into me journey. My favorite one is to go over to Ch GPT, and you can ask CGP, what are all the camera view perspectives that I can use for prompts on the mid journey image generator. I send it. There we go, CPT gave us 13 perspectives that we can use. For example, the bird's eye view or the eye level or a high low angle, a close up perspective, panoramic perspective, first person. If you still don't find the one you like, you can just type 20 more. And I send it and it's going to give you 20 more perspectives that you can use. For example, the underwater view or the profile view, that two point perspective, and here it tells you what it actually does. Let's try some perspectives, and I'm actually quite curious about this one, the eye level perspective. I'm going to copy it and I'm going to write a prompt. Here I have person walking towards the Eiffel tower, ma and level. Let's send it. There we go. Here we have our image, and you can see that it's basically at the eye level. It's taking that view perspective. Let's try a different one. I'm going to write the same prompt. But now I want to use the aerial view. Let's see how different it is. Here we go. Now, the perspective is definitely an aerial view, for example, as if you're flying a drone over. Let's try a different one. I'm going to use the same prompt. I'm going to select another view. I'm going to select the dah view, the car dash and there we go. You can see basically that it's like a dash cam view and there's a person walking towards the Eiffel Tower. My recommendation is to go to C GPT and see what kind of perspective it gives you. If you don't find it, you can just keep on asking it for more and more perspectives. Or you can also just Google and try to find other perspectives that other people have used before that you can also use in your own prompts. 12. Lighting and Shadow Conditions: Now we're going to look into different lightning and shadow conditions. If you're not sure what prompts you can use for your lightning conditions, you can either Google and search for prompts online, or you can always go to CGPT you can ask, what are all the lighting and shadow conditions I can use for prompts on the mid journey image generator. I send it, and here we get some terms that you can use. For example, the time of the day, Dawn, morning, noon, afternoon, directional lightning, if you want, for example, back light or front lighting or top down lighting, the quality of the light, the intensity, the temperature, for example, some colors, yellow orange or red tones. If you still don't find the one you like, you can always ask CPT 20 more. And Chagpts going to start writing some terms. For example, the golden hour, blue hour, low key lightning, candle light, strove lighting. Let's try some of these ones. I'm going to write a prompt. Here I have Cohang Watertown in China. This is one of my favorite small Watertowns let's just send it as normal. This is what we get from the prompt. Now I'm just going to copy the text and try a new one. Imagine, and I'm going to use the same place, the Watertown in China. Now I'm going to add a lighting condition, which is golden hour. Let's see what we get. There we go. It now looks like it's the golden hour. You can see a difference between the normal one and the ones with the special lighting conditions. Let's try a different one. Imagine, I'm going to use the same place, but now I'm going to use the Tight lighting. There we go, here we have the Twilight lighting. Then we have the golden hour and the one with no conditions. From here, you can just keep on checking CGPT and looking for other ones that you might like. If you don't find any, you can always keep on asking for more information. I'm going to try one more with the same prompt. But now, I'm going to use festival lights. There we go. This one looks way better. You can see this one has some festival lights and it looks really nice. This is how you change some of the lighting and shadow conditions. If you know what keyword to use, you can use it. If you don't, you can ask CGPT or Google some keywords. 13. Color Palette: Now we're going to generate some images with specific colors using a color palette. In order to get colors, there are two ways. The first one is that US Cage Pit. What are all the color palette styles I can use for prompts on the mid journey image generator, and it will give you some ideas for different colors. If you want more, you can ask for more. But on this case, what I like is using this website, Imprompt com. On the resources section, you have a colors list. Here you get many different colors and you can select them, you can copy them and you can put them on discord to get that specific colors style. Let's try this one, Chrisman Colors I'm going to copy it. I'm going to go to this chord, start a prompt, and I'm going to write a comic book superhero, and then Chrisman colors. Let's send it. There we go. Here we have a comic book superhero using this color, the Chrisman colors right here. Now let's look for a different color. For example, this one, electric colors. We copy it, and then we paste it, enter There we go. We have our comic book superhero, which looks pretty nice. Now let's try a different one and I'm going to select this one, retro coolors Let's put our prompt and send it. Here we go, we have our superhero in retro colors. I think this one looks really good. It fits very well with the superhero style. So in some cases, you might get CGPT to give you good ideas for lightning and shadow conditions, but in other cases for colors, searching on Google might actually give you better results. In this case, I recommend this website. 14. Compositions: Now we're going to look into different image compositions. For this, I'd like to use this website m prompt, and here we have a compositions list. You can see the different type of compositions that you can generate using M journey. Let's try this one, the Anatomy one. Let's copy it right here. Let's send a new prompt. I'm going to use anatomy com then I'm going to write a comic book, Superhero. Let's see what we get. There we go. This one looks really nice. Let's try another one. I'm going to use the same prompt, a comic book superhero. Let's look for a different one. This one, this one looks really nice. Comic strips. Let's copy it and I send it. Okay. And let's check it out. This one looks pretty nice, like a comic book strip. But it doesn't look exactly like this where you have four different images with lines between them. I think you have to do that yourself. But this looks pretty good, definitely. It looks like a comic book superhero. Let's try a different one. Let's go for this one, mobile app UID sign. Let's check it out. There you go. This one looks pretty nice definitely. We have a superhero in an app. Let's look for another one. Let's try the story board. Let's see if it actually works. This one looks interesting. It looks like the front page of a storybook. If you're looking for different type of compositions, in this case, I recommend checking out this website, prompt and copying some of the compositions right here. See which ones you like. 15. Techniques: Now we're going to look at different technique styles. For this, we go over to the website, prompt. On the resources section, there's a mid journey technique list. Here are different ways that you can create your images. You can make them 16 B, eight bit, three D printed, a cave painting, calligraphy images. Let's try an eight bit image. In this case, I want to use my own image and see how this looks like. I'm going to throw in my profile picture, send it. Now I'm going to write the prompt. I'm going to move this over to get the whole link, and I'm going to write eight bit. Let's send it. There we go. We got an eight bit image using my own picture. Let's try a different one. Now I'm going to write a prompt and I'm going to use a Japanese garden. Let's look for a nice composition. I think I like this one, ballpoint pen art, which is basically a pen. Let's see how this looks like. There you go. This looks really nice. It looks a little bit like a ball point type of pen drawn, but it's really well drawn. Let's try another one. I'm going to write a Japanese garden. Let's look for a different technique. This one looks pretty nice. Drop. Let's copy it and paste it. Here we go. It looks nice. Let's check out one different style. It doesn't look exactly like in the picture, but it looks close enough. Let's try this one etched glass, copy and paste. Here we go. If you want to try different techniques, you can check out this website and copy and paste them in your prompt and see if you find something that you like. 16. Textures: Now we're going to use textures into our prompts. For this, we can check out the same website prompt.com and on the resources section, there's a texture list. This way, we can add a texture into our design. Let's try this one chino let's make a new prompt. I'm going to write a superhero. Then I'm going to copy this one and paste it, see how it looks like. Here we go. Superhero has that chino type of texture. I think I can see it. It's there. But let's try a different one. I'm going to use the same prompt, superhero. Let's try this one, Cobblestone. I think that's a very superhero texture. Here we go. We have our superhero made out of cobblestone. This one definitely worked. Especially image two and four. Let's try one more. I'm going to try this one, the Linguini pattern. Let's see how this will come up like. This one is definitely too random. There we go. We have our Linguini pattern superhero. I think it worked really well for image number two and three. But number four looks very Italian restaurant poster. This website is very useful. If you're trying to get some textures and you don't know what kind of texture you want, you can experiment with this texture list. 17. Photography and Camera: Now we're going to use some photography and camera settings. For this, we go over to our website, prompt and on the resources section, there's a mid journey photography proms list. With med journey, you can change the photography settings. For example, you can change the camera type, the aperture, the lens, the quality, the focal length, even like the camera brand. Let's try something simple, and let's try an eight k image In this case, I'm going to use my own picture. I'm going to add it, and my prompt is going to be this picture, reading a newspaper, and I'm going to use eight K. Let's send it. Here we go. We have some pictures of similar to mine, reading a newspaper in eight K. But let's try something more noticeable. For example, you can change the lens type, the exposure, and I'm going to use these settings. 35 MM lens, photorealistic codec film four k, 400 ISO. These are very precise photography settings. Let's send it. Here we go. Now we have something like a code film type of quality. If you're a person that really knows the camera settings and what each specific setting does, you can always try with me journey and see what the differences are. 18. Fashion Aesthetics: Now we're going to modify the fashion aesthetics of characters that we create. For this, we can go over to the website prompt. On the resources section, there's a fashion aesthetics list. You can select any of these ones that you want your character to have. Let's try this one, the heavy metal outfit. This one looks pretty nice. Let's try to send my own image. Now I'm going to create a prompt, and I'm going to write heavy metal outfit. Send it. Here I have my image with a heavy metal outfit. Let's try a different one. Let's try this one, the arcity outfit. I'm going to write a prompt superhero. And Varsity outfit. Here we have something. For example, we have a superhero suit, but it has the Siper for the varsity team. On this one, it looks like a school jacket. Let's try one more. Let's try this one cult outfit. I'm going to write a man wearing an culture outfit, reading the newspaper. There we go. This ones look more precise. He's definitely wearing that kind outfit and reading a newspaper. But on this one, it seems like the newspaper even went all the way through his pants. Feel free to check out this list for different types of fashion outfits that you can put into your images and the characters that you're creating. 19. Images Inspired by Famous Artworks: Now we're going to create images in the styles of particular artists. For this, we go over to prompt and on the resources section, we have the artist list. From here, we can select styles that are related to particular artists. I'm going to try this one, Akira Tiama I'm going to copy it. The first thing I want to try is to send my own picture. I send it. Now I'm going to write a prompt and I'm going to paste my link, and I'm going to write in the style of Akira Toyama. I send it. Here we go. We have our image, and it looks like this, and the Akira Torima style looks like this. There's definitely lots of similarities on this one. Let's let's try another one. Let's do Pavlo Picasso I'm going to write a man eating an apple in the style of Pavlo Picasso. Send it. Here we go. This should be the style of Pablo Picasso. I'm not that familiar with all of his works to be honest. But let's try another one. Let's try this one, Leonardo Da vinci. I'm going to write Homer Simpson, in the style of Leonardo Da vinci. I send it. Here we go. This one definitely looks really nice. Here we have Homer Simpson in a Davinci style. You can play around with the different styles made by famous artists, and you can combine them with your own prompts. 20. Mood: Okay. Now we're going to customize the mood in our images. You can look for different modes on Google, but in this case, I'd like to ask CGPT what are all the mood styles I can use for prompts on the mid journey image generator. I send it. Here we go. We got 11 different modes. If you want more, you can ask CGP for more. But in this case, I want to try the spooky one, and I'm going to write a prompt. Here I have Homer Simpson in the style of Leonardo Da vinci, and I'm going to ride spooky Mode. Spooky mood. I send it. There we go. This one looks horrible. It basically combines all of the three prompts, and the style is very spooky, especially the first image. It's probably going to give me nightmares. I'm going to just delete this one. If you want to delete, you can always right click and you can click on delete message because I don't want to see this one. Let's delete it. Let's try a different one. Let's do the same prompt. I'm going to go for romantic. I'm just going to write romantic mood. I send it, and there we go. I think this one looks pretty romantic, definitely. Let's try another one, and I'm going to use the same prompt. But now I want to write cosmic slash space mood Let's see what we get. There we go. This one looks pretty unique, definitely. I like this one. You can always play around with the mood and combine different prompts. For example, images inspired by famous artists with specific fashion with specific moods. You can just let your imagination run. But in this case, for setting moods, you can always think of a specific mood or you can ask CPT to give you a list of different terms that you can use. 21. Logo Creation: Now we're going to create different logos using mid journey. I'm going to ask CGPTA prompt. What are all the logo types I can use for prompts on the mid Journey image generator. Here we've got nine different logo types, and I think I want more. I'm going to ask 20 more. I have lots of different ideas right here that I can use. The first logo that I want to create is using my own design. I click on upload a file, and I have this Hanran image that I did on my iPad. I'm going to send it. Now I'm going to add it to my prompt. I'm going to write monogram logo, the Hugo Bakery. Here we've got some logos that I look really well. Of course, mid journey actually struggles when it comes to wordings. Now I want to make another prompt. I'm going to use wordmark logo, the Hugo Bakery. Enter here we've got some nice logos. I like the third image right here. Let's try another one hand drawn logo, the Hugo Bakery. Here we got a few more logos. But I think I don't like any of these ones. You have to be careful which logo type you select. Sometimes it will give you what you want. Other times it will not. There's a lot of times where it might fail. Feel free to look at the different logo styles, and you can ask Cat DPT for more ideas. If you know what you want, you can ask me journey, and you might need to keep on refining and refining your prompt in order to get a logo that you really like. 22. Commands List: Now we're going to go over the command list in mid Journey. To write commands with mid journey, it's very simple. You just have to write the slash key, and here you can see all the possible commands. Here you have the mid journey bot and you can ask all of them. So here you have the commands for the mid journey butt, and you can ask all of the commands that say mid journey on the right side. If you scroll down, it goes to built in. These ones are not the med journey commands, only the ones that say mid journey. If you go over to the documents file of mid Journey, here you have a list of the commands, and they all tell you what they do. If you need a refresh or you don't know what a command does, you can always get the information on this website. Let's start with the first command. 23. Settings Commands (/settings, /info, /help): Now we're going to go over some of the most common settings commands with mid journey. The first command that I want to show is the settings command. You get this one by typing the slash key and then typing settings. We can select this one. From here, you can change the settings in your med journey butt. For example, you can change the stylize, the variation mode. Do you want the images to come fast, very fast or slow? One of the best parts is that you can also change the model. By default, you have the highest model, and you can change the previous mid journey models from previous years. For example, you can try mid journey version four version three, two, one, and there's also the Niji model, which is a model designed for creating anime images. Another setting that I really like is the information setting. You can type the slash key and then type info. Here you can see information related to your profile. So I'm going to send it. Here you can see, for example, how many images you have created, how many hours you have spent. You can also look into the relaxed usage, which in this case, it's actually unlimited, but the fast time is not unlimited. Here you can check how much fast time you have remaining. So with my plan, I get 15 hours per month and unlimited relaxed usage. If you're worried that maybe you're about to finish all of your hours, which is probably going to take a very, very long time to finish 15 hours, you can always check out the user info section. Something else that you can do is that you can click on this one, which is go to your feed. And then it will send you over to this website, and then you can click on Visit Site and you can see all of the images that you have previously generated. You can also see the prompt that you used to create them. For example, for this one, I wrote Studyboard a comic book superhero, and a stylize of 250. You can see all of the information right here. The info command is to get more information related to your account. The last setting that I want to show is the ask setting. For this, we type the slash key and type ask and we can select this one, and here we can ask any question to me journey. Basically, questions related to med journey. For example, how do I get the seed of an image? I send it and here's going to tell you how you can get the seed of an image. It also sends you a link to get more information about it. I can click on it, then visit the site, and here I get more information about the seed parameter. These are three of the most useful settings commands that you can use with med journey. 24. Midjourney Models (/settings): Now we're going to look into the different med journey models. To change models, we can go over to the search bar and we're going to type then settings, and we're going to send it. If we go over to this section, we can change the med journey model. We're currently using 5.2, which is the latest model, and it's also a default model. And from here, we can change to previous models, and we can also change to the NIJ model, which is a model for designing anime images. Let's start with the default model, which is the 5.2, and I'm going to ask command. I'm going to say, as street vendor at a crowded marketplace in Egypt, capturing the colors and atmosphere of the place. Let's see what we get. And there we go. Here we have the images of a street vendor at a crowded marketplace, and here we get the results from mid journey using the model 5.2. I think the image looks really good. I'm going to try the exact same prompt, but now I'm going to use a different model. I'm going to go over to 5.1. Now it's selected and I'm going to send the command. Here we go. We now have the image from mid journey 5.1, and you can see that the image quality is a little different. This one looks very, very realistic. Well, this one is starting to lose some of that realistic quality. Now let's try another model. I'm going to go for 5.0, and I'm going to use the same prompt. Here we go. We now have the image, you can see that the quality is not as good as the first image. The first image looks way more realistic. This one is starting to lose that realistic quality. Now let's try another model, and I'm going to go over to version four. Let's try the same prompt. Here we go, we now have version four, and the quality is noticeably different. Now let's try version three, and the same prompt. There we go, here are the results with mid journey version three, and they don't look good anymore. Let's try the version two now. Here are the results and they look even worse. Let's try the first model. Here we go. These are the results from mid journey version one. This one is supposed to have the worst quality. But I think it looks a little better than version two in this case. This one is version one, you can see it right here. Then you have version two right here. Version, version. Now you can start seeing the improvements from version four. Then we have version five, 5.1 and 5.2, which is the most recent one. The most recent one from 5.2 is the one that has the best quality. It's the most realistic one. I'm going to try another one. I'm just going to regenerate it and try to see some other examples. Here are the new images, and I think they look really good. Of course, compared to the first version of me journey, it has gone a very long way. So by going over to settings, you can switch between the different med journey models. However, remember to go back to the model that you want to use because every time you change the settings, it's going to stick with that new model. If you don't want all of your images to come from med Journey model one, you can change the model, and all of the images will now be generated from that model. 25. Niji Models (/settings): Now we're going to use the NG models. For this, we go over to the search bar and we're going to type the dash Key settings, and we're going to send a command. Here we can change models. By default, you have the last version of me journey, which is 5.2, and we're going to go over to NG model. Here we have two models, version five, and version four. Let's start with version five. NG makes images rather designed in an animate type of format. Let's try a new prompt. I'm going to type imagine. Here I have, a man eating an apple walking in the city. I send it. Here are the results, and you can see that the pictures are in an anime format, even though we didn't specify that we wanted an anime picture. That's what you get when you use the NIJ model. Now I want to check out the version four of the Nige model and use the exact same prompt. Let's see what the differences are. Here we go. We now have the G four model, and the images look way more simple. The D five model has a very modern high quality type of anime, the N four model has a more simple type of anime. The difference is very noticeable. Now I'm going to go back to the NG five model, and I want to try a different prompt, and here I have. Man eating an apple, walking in the city, but now I have Bay Naoko Takeuchi, which is the style of an artist. The images should look a little bit like this, like a Sailor Moon type of image. Let's see how it looks like. Let's check it out and the images look quite nice. They're a little similar to the first ones. But you can have a different type of style. Now let's try another one. I have the same one, but now I have realistic hyper realism. Light magenta and light gold, detailed character expressions, manque, captivating gas, stylish costume design. Here I have a lot of different details that I want to add and let's see the results. And here we go. We get some special results. Now it's basically taking all of the information that I gave it and it's giving us these results. You can make them like any other mid journey prompt. Now, let's see, we can also get Manga type of images, which are the books. I have the same prompt, but here I have black and white Manga style. Let's see. And the ones look really nice. These ones look exactly like a manga type of character. I definitely like this style. Let's try a new one. Let's see if you can actually make pencil sketches. I'm going to use the same prompt, but here I have, rough pencil sketch. A man eating an apple, walking in the city, black and white. Let's send it. There we go. These ones look really nice, definitely. I want to make one more comparison using this exact same prompt, but with the mid journey version five and see if we actually get some differences because G is specifically designed for anime, but mid journey five is not. Let's try it with me journey 5.2. There we go. M journey five looks way more realistic, way more professional while N G five looks exactly like something that a normal person would do. So this one looks something that a professional would do like a museum art piece. This one looks like something that a high schooler can make when he's bored in the classroom. If you like this style better, you can always look for G five. If you want something like this, you can use me journey. Now, remember to change the model back to the model that you want to be the default model. I'm going to go back to me journey 5.2. 26. Raw mode (/settings, --style raw): Now we're going to use the style mode to create more accurate images. For this, we can go over to the documentation, and here we have the information on the style parameter. If you're using the default version of mid journey, There's only one parameter that you can accept, which is style raw. But if you're using G five, you can change the style to different types, for example, the acute type, the scenic type, the original type, or a more expressive type. Here we have some examples. If you don't use the row style, as you can see on the prompt here, you basically get a normal image. Here we have ice cream icon, and that's what you get. But if you use the raw style, you get an actual icon as the image. This one is more accurate. Here I have another example, a child's drawing of a cat, without the row style. This is what you get. But with the row style, you actually get a real children drawing of a cat. If you go down to the NG styles, you can see some of the differences. I don't see a big difference here between default and original, but maybe you can play around and see if there's one you like. But there's a bigger difference once it comes to cute style, the expressive style, or the scenic type. But remember, NG five is good for anime. And here are more examples. Default and original, they look very similar. But cute, expressive and scenic look different. If you want to use any of these parameters, just remember this ones. Let's try the first example. I'm going to go over to settings, and here I actually get the option of always using the raw mode. If I selected, all of them are going to use the raw mode right here. But another thing that I can do is that I can just manually type it. I can just type style raw at the end of my prompt. Let's try it with the written format. I'm going to make a prompt, I'm going to write close up man being cool. And now I'm going to copy this style, and I'm going to paste it at the end. Let's send it. You can see right here is using the raw style. Let's see what kind of image we get. Here we go. I think these images look very realistic. Now let's try it again, but without the raw style. I send it and you can see right here, it's not using the row style. You can see that this image looks very different from the raw style image. The rows style image looks very realistic. Well, this one doesn't, but you still get what you asked it to. If you want to create more realistic type of images, you can always use this style. Now let's try the NG five model. I'm going to go over to the settings, and I'm going to select NG I'm going to write the same prompt of a man being cool I'm just going to select the original style, and let's send it. Here we get a picture of a man being cool in the Aime Niji style. Now let's try it again. But I'm going to select the cute style style cute. I send it. Here we go. We now have acute style using the same prompt. Now let's try another one. I'm going to use the same prompt and the style is going to be expressive. Here we go. We now have the same image, but it's now being more expressive. You can notice that this style is a little bit different. Let's try the last style for the NG five and I'm going to select the scenic style. Let's see how it is. Here we go. We now have the scenic type of style, which looks also interesting. If you like any of these styles using G five, you can always write them. But if you're using the mid journey 5.2 model or higher and you want your images to be more realistic, you can always use the raw style. So here you can check out more of the information, and you can see that me journey only accepts one style while the NG five version accepts four different styles. Of course, you can always keep on customizing and adding different things. But this style is just like a specific parameter. I personally like the style raw with me journey version five. But with ij I don't really see the differences that much between these styles. 27. Stylize Types (/settings, --stylize, --s): Now we're going to look at some of the stylized types. Now we're going to use the stylized parameters. If we look into the documentation, it says that the mid journey but has been trained to produce images that favor artistic color composition and forms. What this means is that there's a certain standard of beauty that the mid journey but has. If you use a low stylized number, The images are going to be similar to the prompt, but less artistic. If you use a hide stylized value, the images are going to be very artistic, but less connected to the prompt. There's a standard of beauty that the mid journey but has and the higher you go, the more it will try to apply those standards of beauty. So here you can see the default in the stylize and the range that you can choose. Basically, you can choose a number 0-1 thousand. For example, the first image right here uses a stylize of zero, which means that the results are going to be closely related to the prompt. In this case, a child's drawing of a cat. Then if you use 50, it starts to distort it a little bit, but it's still similar to the prompt. 100, then two 50, 507 50, you can notice that this is no longer a child's drawing, this drawing is very professional. This is not something that a child can easily do. In this case, you're using more of mid journeys, high standards of beauty. You can see that there are two ways to do this. You can either write the stylize, or you can select the function. For example, if we go over to the settings, We can see that we have different types of stylize. If I'm using a low stylize, we're basically using an S 50. If we go to medium, here we do not see the suffix, but it should be the default of 100 right here, which is the default. If we go to high stylize, it will be two 50, and you can see it right here. These are basically the equivalents. If we go to very high, it's 750. Of course, we can also write them manually in our prompts. Now I want to use my first prompt and I'm going to write close up a wedding at a church. By default, since I cannot choose the no stylized function, I'm going to write it manually. I'm going to write two dash lines, and then the letter S. You can also just write two dash lines and stylize. In this case, I can write either stylize or S. I'm going to use the simple one S, and I'm going to write it right here, I'm going to say zero. In this case, there's no stylize applied. Let's send it. Here we go. We have a wedding at a church. Now I'm going to try the same prompt. But in this case, I'm going to select a stylize of 1,000, which is the highest level. We just did the lowest level. Now we're going to check out the highest level. There you go. The differences are huge. In this one, we're getting a lot of what mid Journey considers to be beautiful. On this one, we're just getting some images similar to the prompt, but we're not getting that mid journey like beauty. These are the differences between zero stylize and 1,000, which is the highest mode of stylize. The other numbers, for example, 50, 150 500, 700, they're just numbers in the middle between this one, which is zero and this one, which is 1,000. These are basically standards of beauty. Now I want to try another one. I'm going to go to the settings, and I'm going to select a default one. In this case, I'm going to go for stylize high and once you put it on your settings, all of your prompts are going to use those same stylized settings until you choose to change them again. Now I'm going to just write it and I'm going to use this prompt. Close up a wedding at a church and I send it and it's basically going to use this stylized settings, the high stylize. Here we get our image, you can see that the image looks pretty good, but it's using a stylized number of 250. If we compare it to the 1,000, this one looks very, very different. This one is like a picture in between what you get right here and what you get right here. Let's try another one, let's go for the low stylize, which is 50. Let's try the same prompt, no stylize, like this, and we will be using these settings and I send it. Here we go. We can see a small difference between this image and this one, but I think it's a little bit noticeable. This one looks a little bit more normal. This one looks a little bit more grand, more beautiful. This one is just way too much, and this one is way too little. This is how you can play around with the stylize. You can either select a default one on the settings function, or you can manually type it by typing stylize or S and selecting a number in 0-1 thousand. You can either do this two ways by writing the parameter manually or by using the settings commands. 28. Remix Mode (/settings, /prefer remix): Now we're going to use the remix mode. So here we have the docs for what remix mode is, it basically allows you to change the, some of the parameters, the model versions, or the aspect ratio between variations of your image. If you look on the first step, we send a prompt, then we make variations, then we can remix those variations by changing some of the information. Here we have for example, some pumpkins, and then we switch over to cartoon owls, and it keeps some of that similar image style as the original image. Here we have another example. We start with some pumpkins, then we modify the prompt, the pumpkins change. Then we modify the prompt again, and the pumpkins change once again. We can also change for example from pumpkins to fruit, and it keeps that type of original image style. However, there are only some parameters that can be affected by remix. These are the aspect ratio, the no parameter, the stop parameter, the tile and the video parameter. So there are basically two ways to activate remix, one is by using the settings command, and the other one is by using the preferred remix command. Let's go over to settings. If you want to activate it, you just have to click on this one, remix mode, and now it will be automatically done by default. If you don't want it by default, you can turn it off, or you can just use the normal command. If you type the dash key and you write remix, you can select this one and you send it and you can see now it has been turned off and you can just type it again. And you can just turn it back on. You can either do this by going over to settings and clicking remix mode or by sending the preferred remix command. Now that it's on, I'm going to write a prompt. Here I have. Three boys playing. There we go. We now have our images, and let's say that I like the first image. I'm going to click on variations. Now it will give us the remix prompt form and we can change it. For example, I can change from boys. I can say three girls playing, and it will make variations similar to this image, but using girls. I submit it There we go. We now have three girls playing and it's keeping some of that similarity from the original image. The first image looks very similar to this image. If you want more pictures of the first image, we can just click on it and not change anything. I'm just going to submit. There we go. We now have the three boys playing right here and three girls playing. You can notice that the style is incredibly similar to each other. Now let's play around a little bit. I'm going to go back to the same one. I'm going to say three dogs playing. Let's see what we get. There we go. We have almost the same style of image as this one. But now we have three dogs playing. Three boys playing and three girls playing, and the images are very similar to each other. This remix function works as a way of changing the original prompt to something different and basically make a remix of your image. You can also change some of the parameters, but not all of them. One of the parameters that you can change is the aspect ratio. Let's say for example that I want to go back to I want the Docs image, and I like the second image. I'm going to click on V two, and I'm going to change the aspect ratio, by typing A R and nine by 16, I can submit it. There we go, we have an image of three dogs playing in this particular style. But now our image has a different aspect ratio. It's like a cell phone type of aspect ratio. This is how you can remix your images and get variations similar to the original picture, but changing some of the proms and some of the parameters. If you want this feature off, you can just click on the dk, go to settings. Then click Remix mode to turn it off, or you can just click on the key, then type remix, and this one, prefer remix and enter it. Now remix mode is turned off. If you want it again, you can run the same command once again, or just go over to the settings. 29. Variation Modes (/settings, high/low variation): Now we're going to look into variation modes. If we go over to the mid journey documentation, here we have some information on variations. Variations are basically when you click on the variation number like 1b2b2, B four. You can either select to make the variations very different or not so different. If we select a strong variation, the images will be different. And if we select a subtle variation, they will not be so different. Let's give it a try. I'm going to go over to the settings, and here we can see the variations. For our prompts. For example, if we start with a low variation, and then we send a prompt, for example, a man walking in a city and we send it, we can see that the results are different, but they're not so different from each other. You can see that they're rather similar to each other. If we select a high variation mode and we send the exact same prompt, We can see that the images are still similar, but a little more different. These ones look much more similar to each other, and these ones have a little bit more of a variation. If we select to upscale an image, for example, this one, number four, we can also once again make variations of this image. We can select a strong variation or a subtle variation. I'm going to click on a strong variation, and you can see it similar, but still a little bit different, and I'm going to select a subtle variation. Now, these images are more closely related with each other, and these ones have a little bit more variation. But now let's try my favorite one, which is the very region combined with the remix function. Now I'm going to turn on the remix. I'm going to select this one, prefer remix and I'm going to send it. Now it's turned on. Now I'm going to click on the very region. You get this special box right here where you can change the prompt. For example, I can select the person right here. I'm going to use the Lasso tool actually. Let's try to get this whole person. I think that should be okay. Now I'm going to change from a man to a pizza let's send it over. You can see now it's going to start changing this particular region because I'm changing the region, and I'm also using the remix function to change the specific prompt. There we go. I can check it out. Picture number two failed. Picture number three failed. It looks like an umbrella. Picture number four definitely looks like a pizza, but picture number one wins. Here we have a pizza walking on the street. I'm going to upscale it. There we go. We do have a pizza walking on the street. We were able to make a variation, grab the variation, and use the remix function to change the prompt. Let's try another example. I'm going to check my settings, and I want to use a high variation mode, and my prompt is going to be a man eating an apple. Here we go. Since we're using a high variation mode, the images will be different from each other. I'm going to select image number four. Now I'm going to change the region very region. I still have the remix function turned on so I can change the prompt. I'm going to select the Lasso tool and I'm going to try to select the apple. I'm going to change a man eating an apple to a man eating a banana. And I can send it. Let's see if this is going to work. There we go. If I open it, I can see that the man is eating a banana, but he did not peel the banana. Next time, I might need to be a little bit more precise with my prompt. But it's still basically the same image, and we change the apple to a banana. 30. Speed modes (/settings, Turbo, Fast, Relax): Now we're going to look into the different types of speed modes. For this, we go over to the settings and you can see right here that there are three types of speed modes. We can select that turbo mode, which makes your images come out really fast, but it takes more of your fast hours. Every month, you have a special number of hours if you're on one of the more pro versions, and if you use the turbo mode, it might take them even faster. Then we have the fast mode, which is the normal one if you have fast hours available and you have then the relaxed mode. The relaxed mode doesn't take any of your fast hours. The relax mode is basically unlimited. You can make as many images as you want, and it will not take time away from your fast hours. Let's check out my info, and here it says, how many fast hours I have remaining. I have 12.13 hours remaining. And I'm going to open the turbo mode by going over to the settings. I'm going to select the turbo mode, and I'm going to write a prompt, and I'm going to check how many fast hours does it take. Also, something that I want to do is that I want to set up a timer to count how fast can it generate this image. I'm going to enter and start the timer three to one, and it generated my image in 23 seconds. Using the turbo mode, I got it in 23 seconds. Let's check my info once again. And you can see right here, we went down from 12.13 hours remaining 80% 0.80 9,212.10. Now we're at 80% 0.68. It took away a few percentages. Let's do the mat. It basically took 0.21% of my fast time to generate this image. Now I want to try again and I want to select the fast mode, which is the mode I've been using all the time, and I want to do a different prompt. Let's select this one. Now I'm going to use the timer. Again, I have the timer. It's reset, and I'm going to click on Start 321. This one took 37 seconds to generate the image and the turbo mode took 23 seconds. This one is still fast bod slower than the turbo mode. Let's check out my info again. Now we're at 12.09 hours remaining, and we went down from 12.10. It basically just went 0.01 point down and we went from 80.68% to 80.62%. In the previous turbo mode, we went down by 0.21%. Now, this one basically went down 0.06%. The turbo mode is taking almost 2.5 times from the fast hours to generate an image compared to the fast mode. Now let's check out the relaxed mode. This one doesn't take any of your fast hours, but it is going to be much slower. So I selected and let's throw in a prompt. Now let's check out with the timer, three, two, one. There we go, 1 minute and 21 seconds. So the slow mode took almost 1 minute more than the turbo mode, and the fast mode was 30 seconds. So the slow mode is definitely much slower by almost 1 minute. If we check out our info, we can see that the fast hours are still untouched. These ones are not affected by the slow mode. These are some of the functions that you can change the speed in how you generate images with med journey. For this, you can check out your med journey pricing if you have a basic plan. I think you have about 3 hours a month or maybe a specific number of generations. For example, the basic plan has limited generations of 200 a month. And then three fast hours or something. But the standard plan, which is the one I have, has unlimited, relaxed GPU generations, and 15 hours of fast GPU time. I think the standard plan is quite good, but it depends on how much you plan to use it. 31. Blend Images Command (/blend): Now we're going to use the blend function to blend some images together. For this, we go over to mid journey in discord and we're going to type the slash key, then blend here you can see that we can select some images and it will combine them together. I selected, and here you can either drag and drop or click on your images. I'm going to select image one, and I'm going to select my normal profile photo, for image number two, I'm going to select this one, the Japanese garden. And we cannot add a prompt after. So I'm just going to click on Enter. Here we go. We have my profile image and the Japanese garden on the back. So it was able to merge both images into one. Now, let's try another one. I'm going to click on the slash key blend. I'm going to select this image of a person flying with some balloons. For the second image, I'm going to use the same one, the Japanese garden, and I click on Enter. There we go. We have a Japanese garden on the background and a person flying on some balloons. Now I want to try something different. I'm going to go over to blend. If you click right here, you can see that it says plus four more. You can click on them and you can still modify some of these ones. You can add more images like image three, image four, image five. You can also select that dimensions. Let's try to merge five images. For image one, I will use the balloon image, image two, the Japanese garden. Three, is going to be this scary house, four is going to be this angel, five is going to be this painting from my mom as well. Here I have five images. Let's see what we get. Here we go. We can see that it's taking the hand painted image from the pictures that my mom painted. I do not see any balloons, but it's combining my face with all of the other ones. This is what you can get when you blend images together. 32. Describe Command (/describe): Now we're going to use the describe command. That describe command allows you to upload an image and it generates different prompts based on that image. For that, you have to use this command. Let's go over to mid journey, and I'm going to write the slash key then describe. Now I can upload an image, so I'm going to use my profile image, and then I just send it. Here we go. We get four different prompt suggestions. For example, a young man with a scarf and shirt in front of a painting in the style of dark yellow. It also gives you the aspect ratio. You can go through them. You can always copy them and put them in your own prompts, or you can select them right here. You can do I want the first one, second, third, fourth, or do I want to regenerate all of them. Now let's try to just regenerate and see what we get. There we go. Now we change some of the descriptions right here. What I want to do now is that I just want to click on Imagine All and see what we get. And you can see that it's going to start doing four different jobs, four separate jobs, and here are the prompts. Here we go. I generated four different jobs based on the prompt description. Of course, we get four different images for each. This is how you use that describe function. You upload your image. Mid Journey is going to look at the information and give you some prompt suggestions, and then you can generate those suggestions or ask for new ones. 33. Shorten a Prompt (/shorten): Now we're going to shorten our prompts using mid journey. For this, we go over to this chord and we're going to type the dash key and write shorten. Here we can write our prompt. I'm going to write this huge prompt that I'm not going to read, and I'm just going to click on Enter. Now it basically gives us four different prompts that are now shortened. If you want to use any of them, you can just click on these numbers. You can also click on show details, and it will tell you about the most important tokens in your prompt. Let's say that I want to generate number one. Let's check it out, number one. If there's something that you want to change or add, you can do it here. But in my case, I'm just going to submit the prompt. There we go. We now have our Eiffel Tower prompt. We can do this for other ones. For example, I can try another one. I'm just going to click on Dak shorten, Dan prompt and I send it, and now we have some smaller prompts. I'm just going to go for number four this time. I'm going to add some stuff. In this case, I'm going to add distend galaxy. And submit. There we go. We now have our image. This is how you can shorten prompts and if you want, you can add more stuff into them. 34. Prefer Command (/prefer): Now we're going to use the prefer commands. For this, there are many options. If you go over to Md journey and you type dash, then prefer. We have already seen the remix mode, so we're going to focus on this ones instead. The first one is the prefer option set. You can basically set custom option, and here it tells me the options that I have previously used. Here you have two parts. The first one is the option, which is basically the name that you want to give this set. In this case, I'm going to call it cyber punk. Then here we have one more option, so we have to click on it, and we can put the values. In my case, I'm going to put three values, which are going to be cyberpunk, skyscrapers and chrome colors. Now I click on Enter and the custom option cyberpunk has now been set to the values. Now if I make, for example, a prompt, I can say Japanese garden, To use this custom option, we have to put two dashes, then the name Cyberpunk, and I can send it and it is basically going to use these values that I provided. There we go. We now just said Japanese garden, and by default, it uses our custom option, which are these parts. All you have to write is the name in this case, Cyberpunk, and it will give you those settings. Let's try another example. I'm going to type prefer option set. The name in this case, I'm going to call it prehistoric, which is the option. Now the values are going to be prehistoric era, dinosaurs, and then an aspect ratio of nine by 16. Now I click on Enter and they have now been set. Now I can use this name. Now I can try it again, Japanese garden, then prehistoric. And I can send it. You will see now that prehistoric was now converted to prehistoric era dinosaurs and the aspect ratio that we selected. Here we go. We now have our Japanese garden in a prehistoric era with dinosaurs. If you want to see the options that you have created, you can go over and type the dash key, then write prefer you can click on this one, prefer option list. You send it, and here you can see all of the custom options that you have created. If you want to delete any of these ones, you can just type the dash Key, then prefer. Go over to this one, prefer option set. Then instead of creating a new option set, we're going to select a previous one. I'm going to go for a Cyberpunk, Okay. And on the values option, we're just going to leave it blank. I click on Enter, and here we have custom option cyberpunk has been removed. Let's try it again, dash prefer prefer option set. I want to delete the prehistoric one and I do not write anything on the values option. I'll just click on Enter now the custom option has been removed. Another thing that you can do is that you can click on the dh Key, then prefer, we're going to go over to this one, which it automatically adds any values to the end of every prompt. Anything you write here, you're going to put it automatically after every prompt that you use. I'm going to click on this one. And it basically gives us one option, which is a new values, so I click on this, and the new value in this case is going to be retro colors. Now I can send it every time we add a new prompt, for example, I click on imagine Japanese garden, and I send it, it will add this at the end. We have the Japanese garden and retro coolors Here we go. We now have our Japanese garden in retro colors. After anything, for example, I can type just a simple house. I send it. Is going to add it at the end of house and then retro colors. There we go, here we have our house. However, this option only allows you to put one thing. If I try it again, I go to prefer suffix and I click on it. I can add a new value. In this case, I'm going to change the aspect ratio. By using AR nine 16, when I send it, it's no longer going to use the previous one, which was retrocolors. Now it's going to use this one, which is the aspect ratio. I can type, for example, Japanese garden again. I send it, and now it's going to aspect ratio at the end. And there we go. We now have our Japanese garden in a phone style type of aspect ratio. This can be very useful if you want all of your images to have the same aspect ratio all the time after every prompt that you use. If you want to know what your current suffix is, you can just go over to slash then settings. Here you'll be able to see the current suffix that you're using. If you want to delete it, you can just click on preferred. Suffix, and we're just going to leave it blank. I click on enter it says now that the suffix has been removed, and you can check by going over to the settings, and now there is no current suffix. Another thing that you can do if you find this useful is that you can click on, then you can select this one prefer auto DM, which basically every time you send a prompt is going to DMU those prompts to the mid journey butt. For example, if I turn it on, Right now, I'm in a specific server. This can be very useful, for example, if you're in a public server and you still want to get your generations sent over to you. If I type, for example, a man saying hello is going to complete the task in whatever server that you used it, and it's also going to send it to you by DM. If I click here, I'm also able to see it is basically sending me a DM with the information. But in my case, I'm just going to turn it off. I prefer Atom, and I send it once again. Now it's disabled. This can be very useful in some scenarios. The last one that I want to show is if you type, then prefer, we have this option which is prefer variability. This one basically changes between the low and the high mode of variations. If I click on settings, Here we have high variation mode, and if I type that, prefer variability and I send it. Now, high variation mode has been turned off. If we check on the settings, It has now been switched over to the low variation mode. If you want to go back, you just type the same prefer prefer variability, we send it. Now the high variation mode has been turned on. If I type on settings, we have been switched back to the high variation mode. This is a simple trick. If you don't want to go over to the settings and do it manually, you can always use this shortcut. 35. Show Command (/show) Get Job Id + Seed Number: Now we're going to use the show command, and we're going to learn how to get the job ID and the seed from an image. For this command, we're going to click on the key and write show. Here, we can get previous jobs that we have generated. For example, for this image, if you want to get the job ID, you can either do it one of these two ways. The first one is to click on the image or to go over to the more settings. And then we can click on Add a reaction. Then we go over to view more, and we're going to search for envelope. And we want to click on the first one. So we send the reaction, and then it's basically going to send us this image over to our mid Journey butt as a DM. We can click on the Md Journey butt, here we can see the job ID and the seed from our image. But my favorite way to do this is just to right click on the image. Then I go over to apps I click on this one, DM results. I click on it and it sends us this image to our DMs. This one is the easiest way in my opinion. So now if we grab this job ID, and we can open, for example, a new server. Let's try a different one. We can click on then show. Now it's going to ask us for a Job ID, so now we can paste it and enter. And it is going to regenerate it. There we go. It now brings back the job over to this new part. This can be very useful if you have a lot of images and you want to put them, for example, in an Excel worksheet where you have all of the jobs next to each other. You can always do it this way. You can have the image and the job ID so that you're able to find it once again. We also get the seed number. The way its work is that you can get images that are somewhat closely related to the original image. They're not going to be exactly related, but somewhat related. For example, if I grab the seed, then I write a new prompt, saying, woman waving hello, then I write and I put the sat number. We now get our image, and you can see that these two images are somewhat similar. For example, the clothing style is the ancient style on both images. This style is very similar. This is how the sat works is going to try to create similar images. But in my opinion, I think that the remix function is much better. 36. What are Parameters: Now we're going to go over into the parameters section. If we go over to our mid journey document, here we have a section where we have the parameters list. Parameters are added at the end of a print and it changes how an image is generated. For example, you can change the aspect ratio. You can remove the sky. You can choose it to stop at a specific generation point. On this part, you have access to all of the parameters, and I recommend checking it out and reading what each parameter does. 37. Aspect Ratio Parameter (--aspect or --ar): Now we're going to look at the aspect ratio parameter. If we go over to the mid journey documents, here we have a section related to aspect ratios. The way that you change the aspect ratio of an image is to add either this parameter or this one. You can say aspect or A R, and then you can select the type of aspect ratio. Here you have some of the information. For example, if you're using version five, you can use any aspect ratio that you want, as well as for N G five. But for version four, you're a little bit more limited. Here you have a list of how the aspect ratios will look like, and you can either select any of these ones. You can use this ones for reference and see if there's one that you like. Here you get a little bit more information. For example, you have the default aspect ratio, which one is a common frame and print ratio, which one is good for print photography, and which one is similar to HD TV screens and smartphone screens. Here it shows you how you can add an aspect ratio to your prompt. Now I'm going to write a prompt. Here I have imagine, the prompt is going to be a space station. To get the aspect ratio, I'm going to write A R or just aspect and the aspect ratio. In my case, I like 16 by nine, which is like a YouTube video type of aspect ratio. I'm going to send it. Here we go. We now have our images with the aspect ratio similar to YouTube videos. If you want the aspect ratio that is for the cell phone, we can type the same prompt. Imagine a space station. And now I'm going to write aspect, and I'm going to change these numbers nine by 16 instead of 16 by nine, and I'm going to send it. Here we go. We now have our space stations, but now the aspect ratio is different. It's like a cell phone size. We can also try another aspect ratio. In this case, I'm going to shorten it. I'm going to say A R let's do one by one, which is a square. I send it. There we go. We now have our image. And something else that you can do if you like these images, but you would like them in a different aspect ratio. You can get the seed from these images and try to generate some similar images. For this, I'm going to right click it, then go to apps, DM results. Then in our DMs, we have the mid journey butt. I'm going to copy the seed number. Then I'm going to go back and I'm going to click on imagine. I'm going to write the exact same prompt. Then I'm going to type then paste the number, then aspect for aspect ratio. I'm going to go for nine 16. I should generate some similar images to this one, but with a different aspect ratio because I'm using the seed. There we go. We now have these images that are somewhat similar to this ones. For example, if you look at this space strip is like a circle with a line, which is very similar to this one, circle with a line on the top. But they're not exactly the same, however, that is the only disadvantage. But the images are somewhat similar to boat. They maintain some of that similarity. 38. Character Reference (--cref) Images with Consistent Characters : Now we're going to use the character reference parameter to create images with consistent faces. For this, we're going to use the CRF parameter, and then we're going to put the URL of the picture that we want the image to have the same character. Here we can see one example. This one is the URL, which is blue hair guide PNG. And in this case, we're going to put the prompt CF and then the URL, and we're going to get images with a similar phase as this one. Let's give it a try. I'm going to use a prompt. For this, I have to use an AI generated image. I'm going to say handsome man and I send it over. Here we have our pictures, but I actually want them to look more realistic. I'm going to go over to the settings. And I'm going to turn on the raw mode. Now I can type the prompt again, handsome man and I send it over. There we go. Our images look more realistic now. I'm going to use the character on number four picture number four. I'm going to upscale it. Now we can put a new prompt. For example, I can say, man playing tennis. Then I type for character reference, and I drag in this URL. I basically just drag in from the image and it's going to paste it and I send it There we go, we get some new images with the same face as this man. This is the original one, and we get a few new ones. We can also put our man in different scenarios. For example, I can type. Imagine a firefighter. Then I'm going to say full body. Then I'm going to type character reference and bring over the URL. Then I enter. There we go. Now we have him as a firefighter. Now let's try another one. I'm going to say a policeman with a police uniform. Then I'm going to write CF for character reference and bring over the URL and enter. While we wait, I'm going to make another one, and I'm going to say superhero Villain, then and let's bring it over. There we go. We have our image where our man is now a police officer, but we cannot fully see the uniform. I think I have to change the prompts a little bit, here we have our superhero villain. This ones look pretty cool. This one also works with the NIJ mode. If you want to create some anime style images, we can move over to the Nj model. Let's try it again. I'm going to say a superhero villain. Okay. Let's bring over the URL. Let's see what we get. There we go. We have our anime looking character. This one can be very useful if you want to create, for example, like an anime book or a comic book, and you want the same character to be in different scenes. For that, you can always use the parameter. 39. Style Reference (--sref) Images with Similar Styles: Now we're going to use the style reference parameter to create images that have the same style. Here we have the style reference information, and this one works very similar to the CRF parameter. You write SRF for style reference, and then the URL right next to it. For example, here we have the image with the URL. If you just say a cat, you get images like this. But if you use the sf parameter, the images are going to maintain that style. Let's give it a try. I'm going to get my image, and I'm going to say, imagine two men playing tennis. There we go. I think the ones don't look too good. They look like paintings. Once again, I'm going to change the settings, and I'm going to turn on the raw mode. Let's use the same prompt. Two men playing tennis. There we go, we got our images. Now let's see which style do I like. They all look very similar though. But let's go for image number one. Let's upscale it. The image looks like this. I think it looks very realistic. Let's try a new prompt, and I'm going to say a soccer team. Then SRF for style reference. Let's bring over the URL by just dragging and I send it. There we go, we now got our images. Which look realistic. They all have the same style as this one, the same type of colors, contrasts, everything basically. Now, let's try another one. But I'm going to switch over to the NG model. Let's select NG and my prompt is going to be a man walking the streets of Osaka in Japan. There we go. We now have our anime style images. From this one, I let's go for number four. I'm going to upscale it. Now let's try a new prompt and I'm going to say a group of kids playing basketball in New York. Then SRF for style reference. Let's bring over the URL and send it. There we go. We now have our images which look very interesting, but now they're made in New York, we changed the setting. But we maintain the same style, and that's what I like. The next type, I'm going to go back to settings. Let's go back to mid journey. And I'm going to use this prompt. Imagine avoid drinking a bottle of apple juice in school. But then I have flat icons. I want icons cute and a white background. Let's send it over. This way, you can also generate icons. There we go. We now have our boy drinking apple juice. Let's see. I think I like the icon number two. I'm going to upscale it. Now I'm going to use a new prompt. I'm going to say a girl walking in Paris near the Eiffel Tower. Then SF let's bring over the URL and send it. There we go. We now have our icons. You can see that it maintains some of those similarities. We can see the girl right here walking in Paris, if we go back to our original icon, it looks like this. If we want, we can always change the prompt. If we want this to be much more similar to this one, we can also keep the rest of the prompt, where I say flat icon skewed white background. Let's try it again. A girl walking in Paris, then flat icon skewed white background, then craft let's bring over the picture and send it over. There we go. We now have our flat icons which look similar to the original icon, which is this one. This ones look pretty nice, but I think there was a small mistake because I asked for a girl and it gave me a boy. In this case, it didn't fully get the prompt correctly, but I think I like the icons. This is how you can create images that maintain a similar style. 40. Chaos Parameter (--chaos or --c): Now we're going to use the chaos parameter. If we go to the mid journey documentation, here we have the parameter for chaos, and we can start it by writing chaos or just C. Basically, this parameter influences how different the images on an image grid. If you have a high chaos, the images are going to be very different and you might have unexpected results, more unusual results. But if you have a low chaos, the images are going to be more similar. Here are some examples. This one is if you're using no chaos values, which is zero, then if you're using a chaos value of ten, 102550100. You can see that with 100, the images are all very different. If you go back to zero, the images are somewhat similar. You can choose a parameter value 0-100. Let's give it a try. I'm going to write a prompt, and I'm going to use chaos value of zero. I type C or chaos, then I'm just going to go for zero, which are going to try to make very similar images. I click Enter, and there we go, we can see that our images are very similar from each other. Now if I do the opposite, I'm going to use the same prompt. But I'm going to use a chaos number of 100, which will make them very different. I click on Enter. Here we go. Here we get the results, and they're very different from each other. Totally different and I cannot even see the medieval castle or the mythical beasts. That's one of the issues with a very high chaos, as it says right here, it will produce more unusual and unexpected results and compositions. These results are very, very unexpected. Let's try a different one. Let's try somewhere in the middle. I'm going to go for a chaos number of 50. There we go. We can see that now we do have a medieval castle and the images are different from each other, but not so radically different as these images with a chaos level of 100. Let's try the last one. I'm going to use a chaos number of 25. There we go. I think I like the chaos number 25 better. The castles look better and the images are not so different from each other. You can play around with the chaos levels if you're not able to find the image that you want. 41. Image Weight Parameter (--iw): Now we're going to use image weight parameters. The way you add an image weight parameter is that you type dW, and then a number. If the number is closer to zero, mid journey is going to prioritize the prompt over the image that you added. If the number is closer to two, is going to prioritize the image over the prompt. So here's an image. In this case, the weight parameter is 0.5. It's going to prioritize the text prompt over the image. In this one, the image weight is two, it's going to prioritize the image over the text prompt. The closer you get to zero, the more it prioritizes the text, the closer you get to two, it prioritizes the image. Let's try an example. I'm going to first add an image, I upload a file, I'm going to upload this one, which are some fruits. I send it. There we go. We now added our image. Now I'm going to add a prompt by typing imagine Now I'm going to add this image URL right here by dragging it. Then I'm going to use a prompt. My prompt is going to be a pencil drawing of a castle. Let's start with an image weight of zero. You type dW, and then the number. By using zero is going to prioritize the text over the image. Let's see what we get. There we go, we can see a huge castle, a pencil drawn castle, and some fruits. The attention is focused more on the castle. Now, let's try the same one. So I'm going to add a prompt. Copy the image URL, and add this one. The pencil drawing castle, but now the image weight is going to be two. Now it's going to prioritize the image above the text, what we called for the text. There you go. It's now going to prioritize the fruits over the pencil drawing or basically whatever we put as the text prompt. On the first one because we used the value of zero, it prioritizes the text, which in this case is going to be the drawing, and on this one, it's going to prioritize the fruits, whatever image we used. And we can also look for something in the middle, for example. Let's try a prompt. We're going to add our image, and let's paste the prompt. I'm going to use an image weight of one. One is rather neutral. You might see both at the same level. Let's see what we get. There we go, by using a level of one, we're rather getting something more neutral. We get a castle and some fruits. But the boat can prioritized at the same level. If we use number two, it's going to prioritize the image, whatever image we added. If we use zero, it's going to prioritize the text prompt over the image. If we use one, it's going to be rather neutral between both. Of course, you can go over to the weight parameters and try other ones. For example, 0.75 or 1205, 1.5, 1705, depending on what you want to prioritize either the image or the text. 42. No Parameter (--no): Now we're going to use the parameter. In the med journey documentation, we have this parameter, which is, and you can put whatever you don't want me journey to include in your image. You can add multiple items. For example, you can ask for a type of painting, and then you can ask what to include. For example, here you have a painting and you can say, I don't want any fruit on my painting, and it will do that for you. Using the no parameter is better than typing, for example, don't add fruit because mid journey is going to read fruit, and it might actually give you fruits. Let's try it out. I'm going to use a prompt and my prompt is going to be a painting of fruits. I send it. Here we have our painting of some fruits, and now we can write what we want to delete. Let's start by removing the grapes. I'm going to make a prompt, imagine a painting of fruits, then dash no, and I'm going to write grapes. No grapes, let's send it. Here we go. This is our original image and we can see some grapes. In our new image, there are basically no grapes anymore. We were able to remove them. We can also remove multiple things. If I start a prompt again, I can say, a painting of fruits, no grapes, oranges, apples. Now it's going to remove grapes, oranges and apples at the same time. For this, remember to use a comma and I can click on Enter. There we go. We now have our painting of fruits, and you can see that there are no grapes, no oranges and no apples in our paintings. Something else that you can do, for example, if you like one of these images up here and you want to remove something from the image, you can, for example, turn on the mix mode, and the click enter. Now the remix mode is on. I can select for example, the image number one and I can upscale it, there we go. I can now click for example, very subtle Now because I'm using the remix prompt, I can modify the original prompt, so I can type no grapes, oranges, apples. And let's see what we get. We can see that it tried to do its best, for example, in this image, I removed all of the apples, oranges, and grapes. But in some images, I can still see some grapes right here, and I think this one is an apple. I think I can see some grapes also right here. In some cases, it might not be the best, but it's still an improvement, but it's not perfect. If you use the remix mode, you might not get fully the results that you want. But if you start a new prompt, it will actually work. That's something that you can keep in mind. 43. Quality Parameter (--quality or --q): Now we're going to use the quality parameter. On the med journey documentations, you can see the information for the quality parameter. For this, we type quality or d q. It's basically going to change how much time mid journey takes generating an image. If you have a higher quality, it will take longer to generate an image, and it will produce them with more details. If you have a lower value, it will make it faster and maybe not have the same amount of details. However, the quality setting does not impact the image resolution. The default quality value is one, and you can always change it. For example, you can choose 0.20 50.5 or 0.1. Here are the differences. If you use a quality of one, you might get an image like this, 0.5 is like this, and 0.25 is more like roughly faster made image that looks like this. Let's give it a tribe. I'm going to make a prompt and my prompt is going to be a castle illustration, then I type quality. And I'm going to go for 0.25. Let's see what we get. There you go. This one was generated quite fast. Let's try another one. I'm going to use the same prompt. But in this case, my quality is going to be 0.5, and I'm going to send it while I wait, I'm also going to do another one, and I'm going to use the quality of one, and I send it as well. There we go, we have our results. This one is the highest quality, which is using quality number of one, and you can see that there's a lot of detail into these images. Then this one is using 0.5, you can notice that the quality is a little bit less, but it's still quite good. And this one is using 0.25. Here you can tell that the quality is not as good as the other ones. This one seems more like a rushed process. Well, this one, you can tell that it actually took its time to generate these images. Keep in mind that the quality parameter is not going to impact the resolution, but it will give you a different look for your images. It also impacts the amount of GPU minutes. A lower quality takes less GPU minutes than a higher quality, which is the default setting. 44. Repeat Parameter (--repeat or --r): Now we're going to use the repeat parameter. The repeat parameter is very simple. It just allows you to run the same prompt multiple times. If you have a basic subscription, you can repeat the prompt two to four times. If you have a standard subscription to ten, and for the pro and mega subscription, two to 40 times. For that, we're going to use this parameter. Repeat or d r. Let's add a prompt imagine a two story house. Then I can type d r. Space, and then the amount of times that I want it to be repeated. In this case, it's going to be two. Then I click on Enter. Now it says, are you sure you want to imagine two prompts of a two story house, and I'm going to say, yes. Here we go. We have two prompts called a two story house. And here we go, our images have now been generated, and we can see which house we like a little bit more. However, I think these images look similar to each other. Something that you can do is that you can combine it with the chaos parameter to have a more diverse range of options. The chaos parameter is basically the parameter that makes your images look different. If you have a low chaos, the images in the grid will look similar, and if you have a high chaos, the images will look different. Let's try different images. I'm going to write the prompt. A A two story house, then the chaos parameter. So I'm going to go for very high chaos. Let's try 80 and repeat. Okay. I wanted two times. Let's send it over. Are you sure? I'm going to say, yes. There we go. We have our results because we're using a high chaos, all of the images are going to be very different from each other. If there's an image that you like, for example, I like this house, we can get more variations of this one. I'm going to go for number two, I'm going to upscale it. I'm going to turn on the remix mode by going over to this one, prefer remix. I think it's already on. I'm going to click on very strong and here I can change the original prompt. I'm going to say r, then two, and I can submit it. Now it's going to ask us, do you want two more variations? I'm going to get eight images, and I say, yes. There we go. Here we have some variations that are still somewhat related to the original image. If if you think that they are so different, you can always try the sto variations, and I'm just going to repeat it, dash r two more times. Let's submit it. There we go, we now have two variations which gives us eight images similar to the original image. That is how you use the repeat function. 45. Permutation Prompts: Now we're going to use permutation prompts. Permutation prompts allow you to generate different variations of a prompt by using some curly braces. For example, you write your prompt, and then in brackets, you can say, red, green, yellow, and then bird. This way, whatever is inside the brackets are going to be processed separately. You're going to have a red bird, a green bird and a yellow bird. You will get three different prompts. You only write one prompt and it's going to give you three different results. This was one of the examples. You can say an illustration of that in brackets, a pineapple, blueberry, rambutan banana, and then bird. In this case, you're going to get a pineapple bird, a blueberry bird, a banana bird, and these are the results that you're going to get. From one prompt, you can have four different variations. I'm going to write a prompt and here I have a boy with sunglasses and in brackets, I have red, blue and green sunglasses. It's going to create three different jobs. Let's send it and it says, Do you want to imagine three prompts from the template, a boy with some glasses? I'm going to say yes. Now here we have our three prompts, a boy with red some glasses, blue and green some glasses. There we go. We now have our images. Here we have the boy with the green glasses, the one with the blue glasses, and the one with the red glasses. From one prompt, we generated three different works with some differences, some variations. You can get quite creative with this ones. For example, you can say the aspect ratio, and then it brackets, you can put four different aspect ratios, and it will generate four different jobs using those aspect ratios. You can also put your prompt and then put the version that you want to use. For example, you can use the mid journey version five, the knee versions, then a test version, and you'll create three different works as well, using different versions. You can look into the more advanced functions. For example, you can combine them. You can put two parts of the same prompt in brackets. This where you're going to get a red bird in a jungle, then a red bird in a desert, then green bird in jungle green bird in a desert. From one prompt, you get four different variations, or you can just keep on going even deeper and deeper into this. If you have any comas in side a bracket and you don't want it to be a different thing, you can put a backslash before it, so it doesn't take it as another object. For this, I recommend looking into the documents for mid journey. To see if any of these cases will apply to you. But that's basically how you use permutation. You put in brackets what you want and it will generate different works for it. 46. Stop Parameter (--stop): Now we're going to use the stop parameter. The stop parameter allows you to finish a job halfway through the process of generating it. For example, if you use the stop, then you can put a value of ten, it will stop it once the generated process is said 10% or if you want it at 50% stopped, it will do it for you. But of course, it's going to be quite blurry. This is what I mean. If I go over and start a new prompt, I say, a man eating grapes and I send it. You're going to get this message saying waiting to start, and then you will start to see the percentages. Now it's at 0% and now it's at 15%, so I can choose to stop it at 15% if I want, now it's at 31. By using this parameter, I can choose at 1% level, I want my image to stop generating. Of course, it will look like this. You will have this blur. Now, this one is the 100% rendered image. But we can try, for example, the same image, and I'm going to say stop then ten. It's going to stop it at 10% rendering and it's not going to look too good, but let's give it a try. There we go. Now our image has been stopped at 10%, and it basically looks like this. You cannot see the image yet. Let's try another one. I'm going to say da stop maybe at 75% enter here we have our image. It's a little bit more noticeable than the previous one, but you still have some of that blur. And let's try another one, stop, and I'm going to try a 50%. There we go. Now our image was stopped at a 50% of the rendering process, it doesn't look too good, but in some cases, you might want an image like that. In other cases, you won't, but if it's useful for you, you can always give it a try. 47. Tuners (/tune): Now we're going to use a style tuner. We can personalize the images from mid journey using a style tuner. For this one, I recommend checking out the documents from mid journey because this one is a little bit more advanced. There are a few more steps that you have to do. The code for this one is going to be dash then tune. Let's give it a try. Dash then tune we're going to select this one tune Now we can write a prompt, but this is not like a normal prompt. You basically add what you want the images to have. For example, I'm going to say that I want the image to be retro elegant, and futuristic. Now I'm going to click Enter and it's going to give me different type of images. For this one, you can decide right here. If you want 16 images, you actually have to multiply by two. You will get 32 different images. If you select 32, you will get 64 images. All of these ones, you multiply by two. I'm going to go for the default, which is 32, and you can also select if you want the default mode or the row mode for the images. So this one is going to cost me 0.3 fast hours. But if I select 16 is going to take a little bit less. But let's go for 32 for the example, and I'm going to click on submit. Are you sure? Yes. It will take around 2 minutes for the tuner to be generated. Now we just have to wait. There we go, our tuner is ready. Now I can just click on the link, and it will send us over to this page. In this page, we can select which image do we like? Do we like this one or this one. Then this one or this one, if you don't like any, just leave it on the center square. But I don't like this type. I prefer going over here and selecting this one with the grids. Now I can select individually, which type of images I like. In this case, I'm going to go for this one, this one looks nice and futuristic. I think this one looks nice as well. And this one over here. Now it's going to give us a specific style, is going to combine our choices and give us a style code right here. If I add another style, for example, let's say, this one, now the code changes. I'm going to unselected and use this code. For this, I just click on copy. Now I can go back to this and I'm going to click on the dash key imagine I'm going to go for a panda bear driving a car. Then I click on style. Then I can just paste it right here. Paste this style, and I'm going to send it. This is what we get. It's basically combining all of these styles that we selected and giving us a specific style, which in this case, it looks like this. If you want to keep the futuristic mode, we can go back and copy the whole prompt, which includes the retro elegant futurism prompts. I can copy this and let's give it a dragon paste. Imagine paste. Then I'm going to use the panda. Let's try it again. There we go. Now we have the retro and elegant and futuristic style, which is a strange combination bodies. If you want, you can also unselect some of these images and just use one and it will have that style. For example, I'm going to go for this one right here. I'm going to copy it and I'm going to use it in a prompt. Now we have the style number right here. Let's use the Panda. Copy and send. There we go. This is our image, and it looks very similar to this image that we selected. This is what the style tuner does. It gives you a lot of different images, and you can select the style that you want, or you can combine different styles. It's going to give you the code, and then you can input that code into mid journey to generate similar images. 48. Random Style (--random): Now we're going to use random styles. The style parameter also has a random function, so you can use this function, and then you can select if you want random 16 64, 32, 1208, which refers to some of these types of tunor styles. It's basically going to generate some random tunor styles and apply it to your image. You can use any of these codes, and you can also add another parameter at the end, which gives you a specific percentage of image pairs selected. So for example, if you use 15, is going to use 15% of these images. If you use for example, 80, it will use 80% of image pairs selected. Let's give it a try. Let's use our panda bear, and let's write style random. Let's use the normal random style first. You can see right here, it's basically selecting a very random style. I give you the code right here. And of course, you can always copy this code if you want another similar style. There we go. This one is a very specific style. Let's try it again. I'm just going to copy this, and I'm going to use the same one random style and send it. Now it's using a different code right here. We're using the same prompt. We just use a random style, so it creates random numbers here. Here we go. We have our image with a new specific style. Now let's try it again. In this case, I'm going to use random 1208, which is going to use images from the 1208 tuner. Let's send it. There we go, we got another image. Now let's try it again. I'm going to write my prompt style. Okay. I'm going to use this style. Random 1208 80% of the image pairs. I send it and here we get our image. Anytime you use, for example, the random, 16, 32, 1208, or 1208 with 15%. The reality is that you never really know what you're going to get. Whenever you use random, it's all going to be very random. It's impossible to know what kind of style you're really going to get from it. In this case, I'm not really sure what the difference is between using 16, 64, 1208, and the specific pairs because everything is going to be randomized at the end. 49. Tile Patterns Parameter (--tile): Now we're going to learn the tile parameter. The tile parameter generates images that can be used in repeating tiles, for example, to create fabrics or wall papers or textures. These are some of the examples. This one is the image, and if you keep on repeating it, the image basically becomes bigger. It gives you images that can be repeated when you combine them together. For this is very simple. You write your prompt, then you write d d tile, and your image will become a tile like this one. Let's start with something simple. My prompt is going to be blue and white lines. Then I write d tile, and I send it. There we go, we have our image, and I'm going to select the first one. I upscale it. There we go. Now we can download it by clicking opening Browser, then right click and save image. Then save. Now we can go over to this website called phong.com checker. It's basically a seamless texture checker, and we're going to add our image. Select the image that we created. You can see we can also sum out and sum in. The image is actually seamless. There's no squares that are different. It's basically a seamless image, and it just keeps on going and going. Let's try another one. Now my prompt is going to be video games, d d tile. There we go. We have some images, and I really like all of these ones. It's hard to choose which one I like the most, but let's go for number three. We upscale it. I click on it, then opening browser, we save it, enter. Now if we go back to our text checker, we can add this file. There we go. Our image is a seamless image of video games. I think this one looks pretty good. It's hard to tell where the lines end. But let's try another one I'm going to do cute tile. Let's see which image we like. I think number three. I click on it, upscale it. Then we open in browser, save the image. Go back to our text checker file, and we click on it. There we go, we have a lot of seamless cats. This one looks pretty nice. If you like it like this, you can always click on download, and it will download this image as a PNG. Now it looks like this, which I think looks pretty nice. If you want to more, you can keep on up scaling it back, maybe like this, and then download. We now have an image, which is much, much bigger, but a little bit harder to see. That's how you can create different tiles from a mid journey image. This can be very helpful if you're doing fabrics or wall papers or textures. Okay. 50. Video Parameter (--video): Now we're going to use the video parameter. The video parameter creates a short movie of your initial image grid being generated. Then if you react with an envelope, the mid journey but will send you a link through direct message. Let's give it a try. I'm going to write a prompt. My prompt is going to be drawing of a cat playing with a ball. For the video to come out, we have to write d d video. And I send it. There we go. We now have our image generated. If you want the video, you can either do this two ways. The first one is to right click on it, then go to add reaction, Vemore then search for envelope and click on this envelope. Now the mid journey but will send you a direct message with the link to the video. So now you can just click on it and see the video. But my favorite way is to just right click, then go over to apps, then DM results, and it will do the same thing. Now we have our message, we can click on it, and we have the video once again. If you want to see the video, you can just click on the link right here. And then visit the site. Here we have a video of our image being generated. Then if you want it, you can just click on the three little dots right here and download it. This might be useful in some scenarios where maybe you have a YouTube channel and you want that effect of your mid journey image being generated. That's how you use the video parameter. 51. Weird Parameter (--weird or --w): Now we're going to use the weird parameter. The weird parameter is like an experimental parameter where you add weird, and then you select the number 0-3 thousand and the lower the number, the less weird the image will be. The higher you go, the weird the image becomes. Let's give it a try. My prompt will be a man eating a potato. Then I type weird or W. I'm going to go for zero first, so there's no weirdness in this image. I'm going to send it. I'm going to copy this, try it again. But now the weirdness is going to be 3,000, the most weird image. I'm going to send it, and let's wait for the results. There we go. This one is our image using zero weirdness, but I think it still looks quite weird and this one using 3,000 weirdness. I think some of these images look a little weird. Let's try it again with a lower number. For example, 500. Then let's also do 301,000. Let's see the results. There we go. Here we have our images. This one is the weird 1,000. Some of these ones look pretty weird, especially the last one. Then we go a little bit less weird 300. It's still weird but not so weird and weird 500. These ones quite weird. Feel free to play around with this parameter to make some images. 52. IMI Prompt Builder: Okay. Now we're going to use a prompt builder from my Prompt. Okay. For this, we go over to prompt.com slash builder? Here we get a builder, and you can see that there are many different elements that we can select to design our own prompts? My prompt is going to be a bear driving a car. If I click on aspect ratio, I can select the aspect ratio that I want. I'm going to do nine by 16, then I click on Enter. Do I want it to be repeated? No. What is the chaos level that I want? How about the image weight? Then the stylize, do I have any specific seed that I want to use Do I have a title? Do I want it anime style? What quality do I want? How about the uplight or do I want it to stop at a specific rendering period? You can play around with these parameters. The part that I like is this one, which is that you can select some elements. So these ones are different types of errors, and you can click on basic elements or digital art elements, photography elements, fashion elements, and here you get even more settings. If we go to basic elements, we can select errors, textures, mediums, colors, themes, materials, techniques, and you can modify the architecture of buildings. You can go for character design. Let's try character design. I'm going to go for evil, the evil design. Then I'm going to go to the basic elements. I'm going to select a time period from the 1,400. Now it's adding it here. Here is my text. To add it here, I have to click on add text. Now it's added to the prompt, and we can keep on selecting different types. For example, the photography type, I'm going to go for types documentary photography. Now it's being added. Then I can select for example camera angles. I'm going to go for an overhead shot. Now it's added here as well. The design of the character, the fashion, so I'm going to go for an academia outfit, apparels. Do I want to add some specific things to my character? Let's go for a ti. Now I have all of these parameters that I have selected, and this is the prompt that we have. Prompt evil 1,400, a bear driving a car, documentary photography, overhead shot, academia outfit, ti and the resolution. If you want to add more parameters, you can select them right here. Now, I'm just going to copy them and go over to this cord, and I'm going to put my prompt and send it. There we go. Here are our images. I recommend using the I prompt if you want to see some examples of how things might look like. If you cannot think of any prompts from the top of your head, you can use this builder to generate some ideas. 53. PromptoMania Prompt Builder: Now we're going to look at another prompt builder, this one from promptomia. For this, we go over to promptomia com mid journey prompt builder, and here we have the builder. You can always set an image prompt if you like. I'm going to give it a try. I'm going to add my own image. I'm going to use my profile photo. If you want to get the link of the image, you can go over to imagine, then just drag the image and get the whole link right here, and then we can paste it to our prompt generator. The whole link. Now in this section, we can add our own prompt. I'm going to write a man driving a car. Here we have this section for the base image. You can select either a beautiful phase as sphere or a landscape. I'm going to go for beautiful symmetrical phase. Then we can add some details. For example, the art medium. Do you want a drawing, and the drawings look like this, and we can click back. Do you want a pencil drawing? They look like this or ink drawing. You can click here to close them. I'm going to try maybe Origami, that would be interesting. Yeah, why not? Let's try the moon game. That looks interesting. Now it has been added to our prompt. But if instead we select for example sphere, we can see that the art medium has now changed. For example, the origami is now related to spheres. If we select landscape, they're actually landscapes now. I'm going to go for the beautiful phase. There's lots of different things you can select, for example, the camera. You can select the film type or the lenses and perspectives. I'm going to go maybe for perspective view. Okay. The camera settings. Would you like a short or a long exposure, the specific shutter speed. You can also change the colors. I'm going to go for colorful, that dimensions, 2d5d multiverse dimensions. There's a lot of different things that you can click here. What I like about it is that it gives you the examples of how it looks like. Let's select the lighting and the types. I'm going to go for neon lamp. You can also choose mimic the style of an artist, and here you have a list of a whole bunch of artists. Okay. So I'm going to go for this one, Lisa Frank. And you can now select the met journey parameters. Would you like it to be a test model, photo realism, the image weight. Let's select the image weight and it gives you some information about what that means. I'm going to go for an image weight of 0.5. There's a whole bunch of different settings right here that you can choose. You can also click here on the sizes and it will give you an aspect ratio. This one is good for films and displays. This one is for mobile images, and some other ones. So I'm going to go for 16 by nine. Now I can just click right here and copy it. Or I can go over here to the top and I can just click on copy. Now let's go over to this chord and put our information. We can see that all of the things we selected have now been added to our prompt, including the image weight and the aspect ratio. We can just send it. Here we get our images, which very interesting. That's how you use this prompt generator. 54. Midjourney Showcase: Now we're going to look at the med journey showcase. If we go over to me journey, here we have the option of showcase. If we click on it, we're able to see some of the images that have been generated by other people. Here you can look at some of the information that they wrote to create that prompt. But if I try to click on an image, I cannot see a lot of the rest of the information. For example, this one, and it doesn't allow me. For that, you have to sign in because it's a feature that only works if you have a paid account. So I'm going to sign in, then sign in with this Cord, then you can do the whole sign up process with your account. So after you sign up with your account, you'll be able to access the mid Journey showcase. Here you can see some of the images that have been created by other people. You can also see the prompt, for example, on this one, if you click on the images going to show you the image. But if you go to the lines, you can also click on copy and then the prompt. However, it's a little strange because on the recent tab, if you click on the image, it sends you to the image. But if you go over to the Explore section, Now, it's a little different. It's a different section. You can click on the images, for example, this one, I will give you more information. For example, here, you can see all of the prompt. You can see the parameters that it used. You can see the style, the size when it was created, and you can also copy it. If you click on the lines right here, you can always copy the prompt as well. You can put it in your favorites, save the image, open it in a new type. There's a lot more features if you click on the Explore page. On the Explore page as well, you can look for other prompts that people have generated. For example, I can say a bear driving, and here we have some of the images generated by other people. You can see which one you like, for example, I think I like this one, so I can click on it, and I'm going to copy the prompt right here. Now I can go over to mid journey, and I can put the exact same prompt and I can just send it There we go. Here I get a very similar image from this one. Of course, is going to be a little bit different, but you can generate similar prompts. You can also click here and look at the parent image. If I click here, I can see the other images that were also generated. If you like any of these ones, you can also go over to the creator's name and you can see some of the other images that the creator has made. The Explore page of the Mid Journey Showcase is very useful if you want to get ideas on how to create your prompts.