Key Takeaways

  • MidJourney now has a user-friendly web interface in alpha development, eliminating need for Discord commands.
  • Initially, users needed 1000 images to access the new interface, now lowered to 100 and soon to be open to all subscribers.
  • The web interface offers easy image generation with settings adjustments, variations, remixed options, and quick access to past creations.

Since it's inception, MidJourney has used the Discord chat platform as an interface for its AI image generation technology. It's not the most elegant solution, but most users have adapted and learned the commands you type in to get the pictures you want. Those times might soon be behind us, as MidJourney now has a web interface in the alpha stages of development.

What Is MidJourney’s Web Interface?

While it may soon fancy, the new web interface for MidJourney is simply a custom website designed to act as the front end of MidJourney. It's user-friendly, designed specifically for MidJourmey, and doesn't require you to learn any special commands or know how to operate a Discord Bot. It also offers a single destination where you can both browse your own creations and those of other users.

This has been long overdue for MidJourney, which has become incredibly popular in spite of the relatively arcane Discord interface it's used so far. if you have access to

Who Gets Access to the New Interface?

When the alpha version of the website was first unveiled, only MidJourney users who had already generated at least 1000 images using the Discord interface were eligible to try it out. Since then, that number has been lowered way down to only 100 images, and I expect that open access to all subscribers won't be long after the publication of this article. Even so, 100 images isn't much, and if you're a new user you'll blow through that halfway through your first month's allocation of images.

Most people who've been subscribed MidJourney for at least one month will already be eligible, so simply head to Midjourney Web and log in using your Discord credentials.

Getting to Know the Web Interface

Assuming that you've got those 100 images in the can, you'll be greeted by the "Explore" page.

The MidJourney web interface.

To get started making your images, switch to the "Create" tab.

MidJourney Create Tab

Here you'll find all of your previously created images, with the prompt, and other useful information about that generation to the right.

The MidJourney image prompt information as displayed in the web interface.

To start a new image generation, type your prompt into the text field marked with "Imagine..." and press Enter.

The MidJourney web interface, showing the 'Imagine' prompt field.

Your latest generated images will appear at the top of the "Create" feed.

Tips and Tricks for Generating Images the New Way

If you already know how to use MidJourney then just about everything you already know translates to the new web interface. The major difference is that you can now see what your options are rather than remembering them or looking them up in a list. The first thing you'll want to do before submitting your prompt is to click on the three horizontal lines to the right of the prompt field. They look like little setting sliders.

MidJourney Settings as seen when expanding the prompt box in the web interface.

Here you can change some important settings with the click of a button:

  • Under "Image Size" you can change the aspect ratio either with a preset button, or with a slider that has all the most common aspect ratios.
  • Under "Aesthetics" you can adjust how stylized, weird, or varied your generation will be.
  • Under "Model" you can pick one of the older MidJourney image generation models, or switch over to the Niji model for anime-style art.

If you click on any of the images that you've created, you'll also see a set of options for that image.

The Midjourney web image option panel.

Here you can:

  • Do subtle or strong variations of the image you've selected.
  • Do a subtle or creative upscale of the image using AI.
  • Remix the image.
  • Pan or zoom the image.
  • Vary one region of the image that you're not happy with, such as a deformed hand.
  • Run the same prompt again.
  • Reuse the image, image style, or prompt as the basis for a new image.

All of these functions were already available using the Discord interface, but now it's easier to use than ever.

The last important tip you'll need is that you can still use most parameter commands in the prompt window. So you can still specify custom aspect ratios or use negative prompts. It really is the best of both worlds.