Image Creation via Midjourney Website: Enjoy All the Tools Available

In this newsletter, read about:

• 🕵️‍♀️ Image Generation via Midjourney Website

• 🗞 News and Top Reads

• 📌 AI Art Tutorial: AI Images with Your Face

• 🎨 Featured Artist: Patrizia Burra

• 🖼 AI-Assisted Artwork of the Week

• 🤓 A Comprehensive Midjourney Guide

🕵️‍♀️ Image Generation via Midjourney Website

Image creation via the website is now available to 5000-club – everyone who has made at least 5000 images on Midjourney. You may check the number of images you created using the /info command in Discord. If you are still below 5000, don’t worry – the feature will be released to everyone over time.

I like the interface for image creation through the website, and today I want to do a brief overview of the key features so that you can use all the tools available on the website once you move there from Discord.

Image Creation

The image creation is currently available on the Midjourney Alpha website – https://alpha.midjourney.com/. The bar for writing prompts is accessible through any website page. But the dedicated page for image creation is the following: https://alpha.midjourney.com/imagine.

Here’s how the interface looks like.

To write your prompt, use the Imagine bar at the top. You may add parameters manually as you do on Discord, or you may click the icon at the end of the Imagine bar to adjust the parameters using the dedicated interface.

As you see, the parameters you can adjust here include:

• Aspect ratio: either choose one between portrait (3:4), square (1:1), and landscape (4:3) or use the slider to choose between other available options.

• Mode: Standard or Raw.

• Version: 6 or any previous version.

• Stylization: --s parameter.

• Weirdness: --weird parameter.

• Variety: --chaos parameter.

• Speed: choose between Relax, Fast, or Turbo modes.

Once you press Enter, you can track the image generation at the Create page.

Variations and Other Tools

When the generation is complete, just click on the image you like, and you’ll see, which tools are available for further image processing.

It’s all the same tools we have been using on Discord:

• Vary Subtle or Strong.

• Upscale Subtle or Creative.

• Remix Subtle or Strong is about the same as Vary but allows you to edit the prompt.

• Pan in any of the four directions.

• Zoom 1.5x or 2x. As far as I see, Custom Zoom is not available on the website.

• Rerun to run the same prompt again.

• Inpaint for Vary (Region).

Additionally, on the website, you can click on Use “Image” to use the corresponding image as a reference for your next generation or Use “Prompt” to copy the corresponding prompt together with parameters to the “Imagine” bar.

Then it’s very convenient that on the Create page, you can see, which images were created originally from the prompt, which are the result of inpainting, variations, upscaling, etc.

More Options

Finally, note that at the right top, you have a few more icons that, among other things, allow you to:

• Copy Prompt, Job ID, Seed, Image, or Image URL.

• Download an image.

• Open it in Discord.

• Search for similar images created on Midjourney.

A few final tricks…

• You may click on the parameters you used if you want to add the same parameter to your next prompt. After the click, you’ll see it being added to the Imagine bar.

• For each generation, you have the options to Rerun, Edit (to change the prompt), Hide, and More (downloading, opening in Discord, reporting, etc.).

I hope this brief guide will make your image-generation process more efficient and comfortable 😀 

Happy Prompting!

🗞 News and Top Reads

• Midjourney released V6 of Niji, a version of their model specifically tuned for Eastern/anime aesthetics.

  • You can enable it by typing /settings and clicking niji model 6 or by typing --niji 6 after your prompts

• Google's research team has recently unveiled Lumiere, an innovative AI model designed for video generation, boasting unparalleled realism and creative flexibility.

  • Lumiere employs a sophisticated "space-time" neural network, enabling the generation of 5-second video clips in a single step, unlike other models that rely on creating frames one by one.

  • This approach results in significantly smoother and more consistent video output, effectively reducing the 'choppiness' often encountered in videos produced by other methods.

  • Capable of generating videos from textual descriptions or animating sections of static images (akin to technologies developed by Pika Labs & Runway), Lumiere's capabilities also extend to video inpainting and style replication as demonstrated in its recent showcase.

• Samsung Galaxy S24 is the first phone released with a new Android feature Circle to Search:

  • To activate Circle to Search, simply press and hold the home button on the screen. Then, use your finger to encircle text or mark an object on the screen that you wish to search for on Google.

  • Circle to Search is also launching on the Pixel 8 and Pixel 8 Pro.

📌 AI Art Tutorial: AI Images with Your Face

In this video, Matt Wolfe shows a new, beginner-friendly tool to inject your face into AI-generated images. It’s also a great way to make images and even videos with consistent characters. The tool has very generous free tier (~ 400 images per month for free), so you can test it thouroughly before deciding if you need a paid version.

🎨 Featured Artist: Patrizia Burra

Patrizia Burra is a professional photographer and digital artist. Her work captures the beauty of the world around us, from the grandeur of nature to the intimate moments of everyday life. Her limited edition prints are highly sought after. Currently, Patrizia is experimenting with AI-assisted art. Check out some of her latest artwork on Instagram @patrizia_burra_photography.

🖼 AI-Assisted Artwork of the Week

🤓 A Comprehensive Midjourney Guide

Share Kiki and Mozart

If you enjoy this newsletter and know someone who might also appreciate it, please feel free to share it with them. Let's spread the word about AI art and introduce more people to this fascinating field!