okuha-logo-small-white

Midjourney AI Art Generator Review – The Best of AI Art?

midjourney-review-feature-image

If you click on a link and make a purchase, I may receive a small commission. As an Amazon affiliate partner, I may earn from qualifying purchases.
Read our disclosure.

Overview

midjourney-discord-interface
Midjourney’s Discord interface.

Midjourney is an innovative text-to-image AI art generator founded by David Holz in 2021 in San Francisco, United States. It has garnered a community of over 13,000,000 Discord members, establishing itself as one of the leading platforms in its field.

Midjourney solely works inside a Discord server, and there’s no need to download and install anything to your computer to use Midjourney.

Midjourney uses large language models (LLMs) and diffusion models to generate random artwork based on the provided text prompts. The quality of the generated images is heavily dependent on the level of detail and complexity within the text prompt.

Check out: Best Midjourney prompts

In terms of pricing, Midjourney offers monthly and yearly subscription options, with slight price differences between them. The Basic plan, priced at $10 per month, excludes the “Fast” image generation mode but is otherwise identical to the Standard plan, which costs $30 per month.

The Pro plan, priced at $60 per month, includes an additional 15 hours of “Fast” image generation compared to the Standard plan. It also offers stealth image generation while maintaining parity with the other plans.

The number of images users can generate per plan varies. The Free plan, although rarely available, offers a relaxed mode. The Basic plan allows for approximately 200 images per month in relax mode. The Standard plan provides around 900 “fast” images and an unlimited number of “slow” images. Lastly, the Pro and Mega plan offers roughly 1,800 and 3,600 “fast” images and unlimited “slow” images.

Overall, Midjourney stands out as one of the best text-to-image (and image-to-image) AI art generator, facilitating the creation of stunning visuals based on text prompts and offering various pricing plans to cater to different user needs.

Features And User Interface

midjourney-user-interface
Midjourney’s Discord user interface.

Midjourney is feature-rich, but only in the form of text prompts and how you can guide the image generation process. Midjourney does not have inpainting capability yet (though the team at Midjourney has mentioned that they are working on it).

The user interface comes with Discord, so you feel at home if you are familiar with how Discord works. However, for newcomers, the Discord user interface is far from ideal for AI image generation.

Midjourney has a lot of parameters that I suggest you get familiar with. It also has an extensive list of different Discord-based commands you can use to either start the image generation process (/imagine) or get more information about your subscription (/info).

You can also use the Midjourney bot on your own Discord server, but it’s only available for subscribers.

The main parameters you should know:

  • --aspect, or --ar: Determines the aspect ratio of the generated image.
  • --no: Negative text prompting, --no flowers prompt tries to remove flowers from the image.
  • --stylize, or --s: Influences how strongly Midjourney’s default aesthetic style is applied to jobs (image generation process).
  • --chaos: Affects how varied the images are in terms of style and aesthetics. Higher values produce more unusual and unexpected generations.

To see how Midjourney is performing and the average wait time for image generation, visit their status page.

Text prompts

Midjourney is mainly a text-to-image AI art generator, but it also has an image-to-image functionality. To start image generation, use the /imagine command in the Discord servers text area:

midjourney-text-prompt-imagine

Example text prompt:

A group of stylized, trendy rubber ducks floats on a serene city fountain, unaware of the looming shadow of a gigantic, playful house cat on the prowl. The ducks, each displaying different contemporary patterns and colors, seem to be in a state of peaceful camaraderie, oblivious to the impending ‘hunt.’ The scene has a comic undertone, evoking laughter and surprise. --ar 3:2 --v 5.1

midjourney-text-prompt-result-example
Example AI image generation results using Midjourney. Image credits.

What sets Midjourney apart from other AI art generators is that you usually don’t need to use a lot of negative prompts with the image generator. Alternative AI art generators tend to need a lot of negative words to fine-tune the end result.

When using parameters, you don’t need to add a comma between the parameters, something you need to do when using other AI art generators.

Image-to-image

Here’s how you can use images in Midjourney:

  1. Fetching online images: Midjourney allows you to fetch any image from the internet to be part of the image generation process. Simply type or paste the web address of an online image with file extensions like .png, .gif, or .jpg. These image addresses are placed at the beginning of the prompt. It’s important to note that prompts must include either two images or one image along with additional text.
  2. Obtaining image URLs: To obtain the URL of an image, you can right-click or long-press on an image and select “Copy Image Address” in most web browsers.
  3. Uploading images to Discord: If you wish to use your own personal images in a prompt, you can upload them to Discord. To do this, click the Plus sign next to the message input area on Discord and choose “Upload a File.” Then, select the desired image from your device to upload it.

By seamlessly integrating online images and personal uploads, Midjourney provides you with the flexibility to enhance your image prompts and generate more diverse and personalized results.

Midjourney versions

midjourney-image-examples
Midjourney design and art aesthetics showcase. Image credits.

Midjourney regularly introduces new model versions to improve efficiency, coherency, and image quality. You can switch between different models by adding the --version or --v parameter to your text prompts or using the /settings command.

Each model has its strengths and specializes in different image types:

  • Model version 5.2 (Midjourney V5.2) offers a new aesthetic system, “high variation” mode, new/shorten command, and new “Zoom Out” and Pan feature (outpainting).
  • Model Version 5.1 (Midjourney V5.1) offers a stronger default aesthetic and excels in coherency. It accurately interprets natural language prompts, produces sharper images, and supports advanced features like repeating patterns.
  • Model Version 5 (Midjourney V5) focuses on generating images that closely match given prompts, particularly in the realm of photography. It may require longer prompts to achieve the desired aesthetic.
  • Model Version 4 (Midjourney V4) introduced a new codebase and AI architecture. It possesses enhanced knowledge of creatures, places, and objects and exhibits high coherency, making it proficient with image prompts.
  • Niji Model Version 5 is a collaboration between Midjourney and Spellbrush, specializing in anime and illustrative styles. It excels in dynamic/action shots and character-focused compositions. Users can fine-tune this model using --style parameters to achieve unique looks, such as cute, scenic, or expressive.

Pricing

midjourney-pricing-240723

Midjourney offers flexible pricing options, allowing you to choose between monthly and yearly subscriptions with slight variations in cost.

Breakdown of the pricing plans:

  • Basic Plan ($10/month): This plan does not include the “Fast” image generation mode but is otherwise the same as the Standard plan.
  • Standard Plan ($30/month): With this plan, you can enjoy approximately 900 “fast” images and an unlimited number of “slow” images.
  • Pro Plan ($60/month): The Pro plan offers all the features of the Standard plan, but with an additional 15 hours of “Fast” image generation. It also includes the option for stealth image generation.
  • Mega Plan ($120/month): The Mega plan gives you the most Fast GPU time while being otherwise the same as the $607month plan.

The number of images that can be generated per plan:

  • Free Plan: The Free plan, although highly unlikely to be available, offers a relax image generation mode.
  • Basic Plan: You can generate around 200 images per month in relax mode.
  • Standard Plan: You can expect to generate approximately 900 “fast” images and an unlimited number of “slow” images.
  • Pro Plan: You can generate roughly 1,800 “fast” images and an unlimited amount of “slow” images.
  • Mega Plan: You can generate roughly 3,600 “fast” images and an unlimited amount of “slow” images.

In terms of efficiency, on average, users can generate 60 images with just one GPU hour. Midjourney does offer commercial licenses for subscribers, but that’s unneeded, as pure AI art can not be copyrighted, and ownership can’t be claimed.

Check out: Can you sell Midjourney art?

Pros And Cons

ProsCons
Best AI art in terms of image quality and aesthetics.User interface. (this will change in the future)
Text prompt precision.User experience. (this will change in the future)
No need to fill the text prompt with multiple negative text prompts.Pricing.
Offers outpainting (Zoom Out and Pan feature).You are limited to the models Midjourney offers.
Occasional mistakes in images (which will cost you credits as they are always counted as one image generation)
Occasional start and end job hiccups.
A table detailing the pros and cons of using Midjourney.

My Experience

“In my experience, Midjourney’s biggest failure is that they reside in Discord. By having that, they are limited to how Discord works, hurting the overall user experience. Midjourney, however, is still one of the best AI art generators out there. The AI art generator produces some of the best AI art, and the art quality makes it a top choice for AI artists”.

Related articles:

How does Midjourney work

What is Midjourney

Midjourney vs. Stable Diffusion

Feature image credits.

Search
artist-profile-picture-avatar

Okuha

Digital Artist

I’m a digital artist who is passionate about anime and manga art. My true artist journey pretty much started with CTRL+Z. When I experienced that and the limitless color choices and the number of tools I could use with art software, I was sold. Drawing digital anime art is the thing that makes me happy among eating cheeseburgers in between veggie meals.

More Posts

Thank You!

Thank you for visiting the page! If you want to build your next creative product business, I suggest you check out Kittl!