Let's get into a few of the apps you can get on your phone right now. I've used a hand­ful of them, and they each have their good and bad points.

Midjourney: The High-Quality Option (With a Catch)

First, let's talk about Mid­jour­ney. This one is known for mak­ing very high-qual­i­­ty, artis­tic, and often real­is­tic images. If you see a real­ly impres­sive AI image online that looks like a fan­ta­sy paint­ing or a detailed char­ac­ter con­cept, there's a good chance it was made with Mid­jour­ney.

But here’s the thing about Mid­jour­ney: it isn't a typ­i­cal app you down­load from the App Store or Google Play. You have to use it through an app called Dis­cord. Dis­cord is a chat app, pop­u­lar with gamers. So, to use Mid­jour­ney, you have to join their Dis­cord serv­er.

Here’s how that works:
1. You down­load Dis­cord and make an account.
2. You join the Mid­jour­ney serv­er.
3. You go into one of the "new­bie" chan­nels, which are chat rooms for new users.
4. To make an image, you type /imagine fol­lowed by your text descrip­tion, which is called a "prompt."

For exam­ple, you’d type: /imagine a photorealistic shot of a cat wearing a tiny astronaut helmet, sitting on the moon.

Then you wait a minute. Your request goes into a queue with every­one else's. The Mid­jour­ney bot will then gen­er­ate four dif­fer­ent image options based on your prompt. You'll see them appear right there in the chat, mixed in with images from oth­er peo­ple. It can be a bit chaot­ic.

If you like one of the four images, you have options to "upscale" it (make a larg­er, more detailed ver­sion) or cre­ate "vari­a­tions" (make four new images sim­i­lar to that one).

The qual­i­ty of Midjourney's images is its biggest strength. It's very good at under­stand­ing artis­tic styles, light­ing, and com­plex ideas. But it has a steep learn­ing curve com­pared to oth­er options. Also, it's not free. You need a sub­scrip­tion to gen­er­ate images after a short tri­al. And because it all hap­pens in a pub­lic chat room, every­one can see what you are cre­at­ing.

DALL‑E 3 (via ChatGPT or Microsoft Copilot): The Easiest to Use

Next up is DALL‑E 3. This is the image gen­er­a­tor made by Ope­nAI, the same com­pa­ny that makes Chat­G­PT. And that’s the main way you use it. If you have a sub­scrip­tion to Chat­G­PT Plus, you have access to DALL‑E 3 right inside the chat.

This makes it incred­i­bly easy to get start­ed. You don't need to learn spe­cial com­mands. You just talk to Chat­G­PT like you nor­mal­ly would.

You can say some­thing like: "Can you cre­ate an image of a red pan­da learn­ing to cook in a bright, sun­ny kitchen? Make it a cute, car­toon style."

What’s real­ly good about this approach is that you can have a con­ver­sa­tion to refine the image. If the first image isn't quite right, you can just tell Chat­G­PT what to change. For exam­ple: "That's great, but can you make the red pan­da wear a chef's hat? And add a bowl of spilled flour on the counter." This back-and-forth makes it very user-friend­­ly.

The qual­i­ty is very good, and it's par­tic­u­lar­ly skilled at under­stand­ing long, detailed prompts and cre­at­ing images with text in them, some­thing many oth­er gen­er­a­tors strug­gle with.

A big advan­tage is that you can also access DALL‑E 3 for free through Microsoft Copi­lot (which was for­mer­ly called Bing Image Cre­ator). You get a cer­tain num­ber of "boosts" for faster gen­er­a­tion, but you can keep mak­ing images even after those run out, it just might be slow­er. The Chat­G­PT app and the Copi­lot app are both avail­able for iPhone and Android.

Leonardo.Ai: A Great All-Rounder with a Free Option

Leonardo.Ai is anoth­er strong option that has become real­ly pop­u­lar. It start­ed as a web tool but now has a mobile app. It strikes a good bal­ance between being easy to use and offer­ing pow­er­ful fea­tures.

One of the best things about Leonar­do is that it gives you a decent num­ber of free cred­its every day. These cred­its reset, so you can keep mak­ing images with­out pay­ing, as long as you don't use it too much in one day.

Here’s how Leonar­do gen­er­al­ly works:
1. You open the app and type a prompt in the text box.
2. You choose a spe­cif­ic mod­el. This is a key fea­ture. Leonar­do has its own gen­er­al mod­els, like "Phoenix" and "Leonar­do Dif­fu­sion," but also has lots of fine-tuned mod­els trained by the com­mu­ni­ty for spe­cif­ic styles. You can find mod­els for cre­at­ing water­col­or paint­ings, vin­tage pho­tos, ani­me char­ac­ters, pix­el art, and more.
3. You can also add a "neg­a­tive prompt." This is where you list things you don't want to see in the image. For exam­ple, if you're mak­ing a pic­ture of a hand and the AI keeps adding a sixth fin­ger, you could add "extra fin­gers" to the neg­a­tive prompt.
4. You hit "Gen­er­ate" and it cre­ates a few images for you to choose from.

Leonar­do also has oth­er tools. There's an "AI Can­vas" that lets you edit your images in more detail, expand­ing the pic­ture, remov­ing things, or adding new ele­ments. This gives you much more con­trol than just typ­ing a prompt and hop­ing for the best. The image qual­i­ty is high, and the abil­i­ty to choose from so many spe­cial­ized mod­els is a huge plus.

Apps That Integrate AI Into Photo Editing

Some apps aren't just for gen­er­at­ing images from scratch. They are full pho­to edi­tors that have added AI fea­tures. Apps like Pic­sart, Fotor, and Can­va fall into this cat­e­go­ry.

With these apps, the AI image gen­er­a­tor is just one tool among many. You can cre­ate an image from a text prompt, and then imme­di­ate­ly take that image and add text, apply fil­ters, com­bine it with oth­er pho­tos, or use it in a social media tem­plate.

Here’s a typ­i­cal process in an app like Can­va:
1. Start a new design project, like an Insta­gram post.
2. Go to the "Ele­ments" sec­tion and find the "AI image gen­er­a­tor" tool (it might be called some­thing like Mag­ic Media).
3. Type your prompt, like "a min­i­mal­ist water­col­or paint­ing of a cac­tus."
4. Choose a style, such as "Pho­to," "Draw­ing," or "3D."
5. The app gen­er­ates the image, and you can drop it right onto your design can­vas.

This is use­ful if you're mak­ing con­tent for social media, mar­ket­ing mate­ri­als, or pre­sen­ta­tions. You don’t have to gen­er­ate an image in one app and then upload it to anoth­er to fin­ish your design. It all hap­pens in one place. The image qual­i­ty might not always be as pure­ly artis­tic as what you'd get from Mid­jour­ney, but the con­ve­nience is a major ben­e­fit. Many of these apps offer a free tier that includes some AI image gen­er­a­tion.

So, which one should you choose?

It depends on what you want to do.

• If you want the absolute high­est qual­i­ty images, espe­cial­ly for artis­tic or fan­ta­sy scenes, and you don't mind learn­ing a quirky inter­face and pay­ing a sub­scrip­tion, Mid­jour­ney is a strong choice.
• If you want the eas­i­est, most con­ver­sa­tion­al expe­ri­ence and the abil­i­ty to refine your ideas by just talk­ing, DALL‑E 3 (through Chat­G­PT or Copi­lot) is the way to go. The free access via Copi­lot makes it a great start­ing point.
• If you want a pow­er­ful tool with lots of con­trol, dif­fer­ent style mod­els, and a gen­er­ous free dai­ly allowance of cred­its, Leonardo.Ai is excel­lent.
• If your main goal is to cre­ate images for social media posts, fly­ers, or oth­er design projects, an all-in-one app like Can­va or Pic­sart is prob­a­bly the most effi­cient choice.