Skip to main content
  1. Home
  2. Computing
  3. Guides

How to use Midjourney to generate AI images

By
Midjourney River image.
Midjourney

The era of AI-generated artwork is upon us, and the internet is filled with users trying to create the perfect prompts to lead AIs to create just the right images – or sometimes, just the wrong ones. Sound like fun? One of the more common AI tools is Midjourney, which people use to create dreamlike landscapes and subjects with just a few words.

Recommended Videos

Difficulty

Easy

Duration

15 minutes

What You Need

  • Discord account

If you’d like to experiment with Midjourney, we’ve got good news: It’s free to sign up, and you can start trying out the AI generator in just a few minutes. Here’s everything you need to know about using it for the first time.

How to start using Midjourney

Step 1: Make sure you have a Discord login. Though there is a dedicated site now, it is invite only. For the vast majority of us, Midjourney works entirely on Discord, so you’ll need an account there to use it. Signing up for Discord is also free if you haven’t done it yet.

Step 2: Visit the Midjourney website. Here, choose Join the beta. This will automatically take you to a Discord invite.

Midjourney Join the Beta.
Image used with permission by copyright holder
Related

Step 3: Accept the Discord invite to Midjourney. Choose to Continue to Discord.

Midjourney Accept Invite.
Image used with permission by copyright holder

Step 4: Your Discord app will automatically open. When it does, select the ship-like Midjourney icon on the left menu.

Step 5: In the Midjourney channels, locate the Newcomer rooms. There will typically be a number of newcomer rooms open, with names like “newbies-108.” You can select any of these to begin.

Finding the newcomer rooms on Midjourney's Discord.
Digital Trends

Step 6: Now you’re ready to begin creating AI art. Before you get started, note that you only have a certain number of prompt options available as part of your free trial. You can create around 25 free images. After that, you’ll have to purchase a full membership to continue. If you would rather not spend any money, it’s a good idea to take some time and think about just what you want to create on Midjourney. If you want, you can type “/help” to get a list of tips to peruse.

Step 7: When ready, type “/imagine” in the Discord chat for your newbies room. This will create a prompt field where you can type the image description. The more precise that you can be with your description, the better the AI will be able to produce good results. Be descriptive, and if there’s a particular style that you are looking for, include it in your description. There are terms of conduct to follow here, but if you keep things clean, you shouldn’t have anything to worry about.

When finished, select Enter to send your prompt.

Midjourney Image Options.
Image used with permission by copyright holder

Step 8: Give Midjourney a minute to generate your images. Typically, the AI will create several different versions based on your description. You now have a number of options to continue.

Look below the images, and you’ll see a section of U and V buttons labeled 1 through 4. The numbers correspond to the four images that Midjourney produced. Choosing U will upscale that particular image into a larger, more defined version. Choosing V will create an all-new image based on the present image that you choose. You will also see a refresh button to the side to request a new set of images. Keep in mind that each of these choices will use up some of your available free prompts, so only do it if you are sure you want to proceed.

Step 9: Once you look at a single image, you'll still have some more options about how to alter it: * Vary — Creates four more images that will look somewhat like the selected image. * Zoom Out — Will shrink the image and generate more context-based imagery around it. * Arrows — Will "pan" the image and fill in newly exposed areas with context-based imagery. * Heart Symbol — Will favorite an image, to allow you to find the image easily in your Midjourney Gallery. * Web — Allows you to open the image directly in your Midjourney Gallery.

Step 10: If you plan on using a lot of Midjourney, you can use any bot channel in Midjourney’s Discord and type “/subscribe.” This will create a link that you can follow to pay for a subscription. Those who are really serious about using Midjourney in the long term will also want to take a look at the manual, which will provide you with a greater list of commands and some advice about how to create images.

For more AI image-generating options, check out what Microsoft is doing in the field, too.

Editors’ Recommendations

Topics
Tyler Lacoma
Tyler Lacoma
Former Digital Trends Contributor
If it can be streamed, voice-activated, made better with an app, or beaten by mashing buttons, Tyler's into it. When he's not…
I tested the future of AI image generation. It’s astoundingly fast.
Imagery generated by HART.

One of the core problems with AI is the notoriously high power and computing demand, especially for tasks such as media generation. On mobile phones, when it comes to running natively, only a handful of pricey devices with powerful silicon can run the feature suite. Even when implemented at scale on cloud, it’s a pricey affair.
Nvidia may have quietly addressed that challenge in partnership with the folks over at the Massachusetts Institute of Technology and Tsinghua University. The team created a hybrid AI image generation tool called HART (hybrid autoregressive transformer) that essentially combines two of the most widely used AI image creation techniques. The result is a blazing fast tool with dramatically lower compute requirement.
Just to give you an idea of just how fast it is, I asked it to create an image of a parrot playing a bass guitar. It returned with the following picture in just about a second. I could barely even follow the progress bar. When I pushed the same prompt before Google’s Imagen 3 model in Gemini, it took roughly 9-10 seconds on a 200 Mbps internet connection.

A massive breakthrough
When AI images first started making waves, the diffusion technique was behind it all, powering products such as OpenAI’s Dall-E image generator, Google’s Imagen, and Stable Diffusion. This method can produce images with an extremely high level of detail. However, it is a multi-step approach to creating AI images, and as a result, it is slow and computationally expensive.
The second approach that has recently gained popularity is auto-regressive models, which essentially work in the same fashion as chatbots and generate images using a pixel prediction technique. It is faster, but also a more error-prone method of creating images using AI.
On-device demo for HART: Efficient Visual Generation with Hybrid Autoregressive Transformer
The team at MIT fused both methods into a single package called HART. It relies on an autoregression model to predict compressed image assets as a discrete token, while a small diffusion model handles the rest to compensate for the quality loss. The overall approach reduces the number of steps involved from over two dozen to eight steps.
The experts behind HART claim that it can “generate images that match or exceed the quality of state-of-the-art diffusion models, but do so about nine times faster.” HART combines an autoregressive model with a 700 million parameter range and a small diffusion model that can handle 37 million parameters.

Read more
ChatGPT app could soon generate AI videos with Sora
Depiction of OpenAI Sora video generator on a phone.

OpenAI released its Sora text-to-video generation tool late in 2024, and expanded it to the European market at the end of February this year. It seems the next avenue for Sora is the ChatGPT app.

According to a TechCrunch report, which cites internal conversations, OpenAI is planning to bring the video creation AI tool to ChatGPT. So far, the video generator has been available only via a web client, and has remained exclusive to paid users.

Read more
Adobe releases its first commercially safe Firefly video generating AI
Firefly video still shot of an Icelandic horse

Following on the success of its IP-friendly Firefly Image model, Adobe announced on Wednesday the beta release of a new Firefly Video model, as well as two subscription packages with which to access its audio and video generating abilities. Generate Video, according to the announcement post, "empowers creative professionals with tools to generate video clips from a text prompt or image, use camera angles to control shots, create professional quality images from 3D sketches, craft atmospheric elements and develop custom motion design elements."

The model will initially be able to generate video in 1080p resolution to start, though the company plans to release a 4k model for professional production work in the near future. Like the image generator, Firefly Video is trained exclusively on Adobe stock, licensed, and public domain content, making its outputs usable in commercial applications without fear of them running afoul of copyright or intellectual property protections. And, unlike Grok 2, there's minimal chance of it outputting racist, offensive, or illegal content.

Read more