Creating your first image

- [Instructor] In the last movie, we became familiar with the community creation section where we can explore image generated content from our community. And if we find something we like, we can click on that image, discover the prompt details, as well as the model and presets that were used to generate that image, not to mention remix it and create our own images from that text prompt. In this movie, we're going to create our own images from scratch. Now, there are two different ways that we can go into the image creation section of Leonardo. One of those is just underneath the AI tools, simply clicking on the image creation section. Now we're in image creation and any images that you have created will show up here on the bottom. And our main area of focus happens to be our ability to type a prompt or use an image as a reference in a number of different ways. Since we're just starting out and we're unsure of what we want, one thing we can do is ask Leonardo to roll the dice or come up with a prompt for us to try. If we come over here to the the little star icon here at the side, simply choose new random prompt, and Leonardo will get to work providing us with a prompt that we can try. This case, an electrifying, ominous, insectoid, fish-like creature, shimmering exoskeleton, cracking with neon colors against a dark, foreboding backdrop. If this sounds interesting to us, we can simply generate this right now, using the default settings or alternatively, go back to that same star icon and choose a different random prompt, which might appeal to the certain aesthetic that we're looking for. A menacingly detailed gnome with grotesque features, every aspect teeming with horror, amidst intricacy, and it goes on. Of course, we don't have to accept this prompt, and at any time, we can select all of the contents and press delete. In fact, let's try out our own prompt. Before we do, I want you to note that I'm on classic mode. Flow state, this tab here, is something brand new to Leonardo, and we're going to get familiar with that in a few videos from now. Click on classic mode if it's not already selected, and let's type in this prompt. A 30-year-old man with green eyes, wearing a baseball cap in downtown Manhattan, shots on a 35 millimeter lens. Now, in terms of prompts, giving detail is going to help generate images that reflect what you're looking for. Whatever you place at the beginning of the prompt is also going to be weighed more heavily than what you place towards the end of the prompt. If you're looking to see how detailed you can get with prompts, one thing you could do after you type something in that's very simple is ask Leonardo to take a look at the prompt and add more detail for you. So in this case of a 30-year-old man with green eyes wearing a baseball cap in downtown Manhattan and shot on a 35 millimeter lens, we'll go back to this little star icon and ask it, in this case, to improve the prompt. We can see that Leonard gets to work and has made a few changes, making the prompt longer, a young man weathered by time and experience gazes intently with emerald eyes beneath a worn baseball cap. Now some of these details, I don't actually want. I don't think I want to have emotion of someone who is weathered by time and experience. I can remove that from the prompt, or alternatively, actually come here to the star icon and ask it to edit it with AI. And in this case, I'm going to type in, make man happy so that he's not so weathered by time and experience. I'll choose done. We can see here that the weathered by time and experience still is listed here, but beams with a warm smile. So in this case, I'll just remove the weathered by time and experience and use, a young man beams with a warm smile, his emerald eyes sparkling with joy beneath a worn baseball cap. So now that we have this prompt, it's time to take a look at the settings that we're going to use to generate this image. One of the most important is the model that we'll be using for the generation, Phoenix 1.0 is the current, default model being used for the generation of these images. There are various models for specific styles. In the case that you want something more cinematic in portrait, you have cinematic keynotes available to you, not to mention portrait perfect. You're looking for something more along the lines of graphic design or concept art. There are models for those styles as well. So I'll choose the cinematic keynote and then just close out the model presets. We'll notice that the style currently being used to create these images is also cinematic. I can choose a few other ones. Maybe I'll choose portrait instead. The contrast for our photos is set to medium, and the only option we have for generation is fast generation. We can change the generation and have more options based on the model that we have, not to mention if we have a paid account. One of the biggest things that will have an impact on your image is the dimensions. By default, it will be set to one by one, but I've changed this to two by three. And then we have the quality of the image generated. Available to us on the free account, we have both small and medium. In order to use less generation credits, I'm going to choose small. I'll leave the number of images being produced as four as that's all I have option to on the free account. And now here in classic mode, with our prompt that we use both AI as well as Leonardo to edit it and make it more detailed, I'll choose to generate these four images. We'll see as these images generate the text prompt that we used here at the top available for us, the model, not to mention the style and the amount of images being generated. So here are the characters generated from our text prompt. And maybe after we see this photo, we want to change this a little bit. So next to each image, we have a few options. One is to, at any point in time, reuse the prompt, which will load it back up here into the text prompt field. The other is to try to edit this with AI. In fact, I'm going to click on this pen icon. We'll edit with AI again, and I'll say, make image from 1930s in black and white at night. Now generate new images based on that information. And you should see that your text prompt was also changed. So Leonardo took in what you entered, and it then added in detail, using AI in order to produce these new images, such as this circa 1930 that you see here, as well as shot on a vintage camera. And now we have something that was shot at night in a few photos, not to mention that black and white 1930s style that we asked for. As we produce this image, if you aren't a free account, you're going to want to keep track of the amount of tokens that are you are using. Right now, I'm already down to 85, based on the generations that I've made. Now this renews every single day, so by tomorrow, you'll have another 150 tokens available. But if you are ideating, you may want to choose a smaller size in order to make sure that you don't use up all your credits. And that is how we can generate images, using Leonardo.AI.