41 Midjourney Tips for Beginners (Short Version)

Future Tech Pilot
2 Apr 202408:28

TLDRThis transcript offers 41 tips for beginners using the Midjourney AI art generator. It covers a range of techniques from using image prompts and blending multiple images to utilizing emojis and character references for unique outputs. Tips include using the 'info' command to track your progress, earning free fast hours by rating images, and experimenting with different aspect ratios for various creative projects. Advanced users are encouraged to explore parameters like chaos and stylize for more control over the generated images. The transcript also suggests using negative prompting to exclude unwanted elements and remixing prompts for consistent art. For those who need help crafting prompts, the transcript recommends using AI like Chat GBT. Finally, it emphasizes the importance of testing different parameters and experimenting with style references and upscales for the most satisfying results.

Takeaways

  • πŸ–ΌοΈ Use an image as a prompt with the parameter D-iw3 to recreate an exact picture.
  • πŸ”„ Blend up to five images at once with the forward SL blend feature on Discord for a unique combination.
  • 🎨 Specify a color in your prompt to generate images using that color as a reference.
  • πŸ˜ƒ Include emojis in your prompt for random and unique image generation; combine emojis for more randomness.
  • ✍️ Write prompts with permutations to save time, but be cautious as it may quickly use up your fast hours.
  • ⭐ Rate images on the Midjourney website to earn free fast hours; top 2,000 raters daily receive a GPU fast hour.
  • πŸ“Š Use the info command on Discord to track the number of prompts and images generated.
  • ⏱️ Spend your fast hours wisely by using the repeat parameter with a small chaos value to explore a prompt's potential.
  • πŸ“ Utilize Smart folders on the website for efficient organization of prompts based on keywords.
  • 🌐 Start with a basic idea in your prompt and then expand upon it to refine the AI's output.
  • πŸ–ŒοΈ Use the very region feature for in-painting, allowing you to replace characters in a scene.

Q & A

  • How can I recreate an exact picture using an image prompt?

    -To recreate an exact picture, use the image as an image prompt and append the image weight parameter D-iw3 at the end of your prompt.

  • What is the forward SL blend feature on Discord, and how can it be used?

    -The forward SL blend feature on Discord allows you to blend up to five images at once without adding words to your image prompt.

  • How can I generate images using a specific color?

    -To generate images with a specific color, use that color as a character reference in your prompt.

  • What role do emojis play in generating random and unique images?

    -Emojis can be used in your prompt to generate random and unique images. Combining emojis can lead to even more randomness.

  • How can I save time when writing prompts?

    -Write prompts with permutations to save time, but be cautious as this may quickly use up your fast hours.

  • What is the benefit of rating images on the mid Journey website?

    -Rating images on the mid Journey website allows you to earn free fast hours. The top 2,000 people who rate images per day will receive one GPU fast hour.

  • How can I explore the full potential of a prompt?

    -Spend your fast hours and use the repeat parameter along with a small chaos value to explore what a prompt has to offer.

  • What is the purpose of Smart folders on the mid Journey website?

    -Smart folders are used for fast organization. You can add any keywords you want, and mid Journey will fill that folder with prompts containing those words.

  • What is a basic prompting tip for starting with a prompt?

    -Start small with your prompt and then elaborate after establishing what mid Journey will create with the basic idea. Then, start inserting your expectations into the prompt.

  • How can I replace a character in a scene using mid Journey?

    -You can replace any character in any scene by using the very region, also known as in-painting.

  • What is the nii model, and how can I use it in my prompts?

    -The nii model is a mid-journey model trained on anime. You can use it by writing D-nii6 at the end of your prompt or by speaking directly to the nii bot.

  • How can I create a realistic picture and then transform it into a surreal result?

    -Create a realistic picture in mid journey and then image prompt it into nii. You might want to remove any words in your prompt that refer to photo realism to achieve surreal results.

Outlines

00:00

🎨 Creative Image Generation Tips

This paragraph provides a comprehensive guide on how to recreate images, blend multiple images, use color references, emojis, and permutations for unique image generation. It also suggests rating images on the Mid Journey website to earn fast hours, using the info command on Discord, and exploring prompts with the repeat parameter and a small chaos value. Tips are given for using Smart folders, starting with basic prompts, and utilizing advanced techniques like inpainting, negative prompting, remixing, and aspect ratio considerations. The paragraph also covers using specific models like D-nii for anime-style images and concludes with advice on troubleshooting and experimenting with various parameters for different styles of images.

05:00

πŸ€– Advanced Techniques and AI-Assisted Creativity

The second paragraph delves into more advanced techniques for creating seamless patterns, using AI to generate prompts, and manipulating the chaos parameter for unique images. It discusses the use of style raw for more realistic images, higher stylized values for beauty, and a combination of stylize, chaos, and weird parameters for fun images. The paragraph also suggests using multi-prompting and style references, especially with high chaos values, to achieve interesting results. It provides tips on locking in seed numbers, creating outlines, and using forward SL in on public Mid Journey channels. The speaker also invites the audience to access the Mid Journey course for a more structured approach to using AI for art.

Mindmap

Keywords

πŸ’‘Image Prompt

An 'Image Prompt' is a specific image used to guide the AI in generating a new image that is similar or inspired by it. In the context of the video, it is suggested to use an image as a prompt with the parameter 'D-iw3' for recreating an exact picture, showcasing how it is central to the process of image generation.

πŸ’‘Forward SL Blend Feature

The 'Forward SL Blend Feature' is a tool that allows users to blend up to five images at once without adding words to the image prompt. This feature is highlighted as a way to combine multiple visual elements into a single image, demonstrating its utility in creating complex and layered visuals.

πŸ’‘Emojis

In the video, 'Emojis' are recommended for use in prompts to generate random and unique images. They can also be combined for additional randomness. Emojis serve as a creative and fun way to introduce elements of chance and diversity into the image generation process.

πŸ’‘Permutations

'Permutations' refer to the different ways in which the elements of a set can be arranged. In the context of the video, writing prompts with permutations is a method to save time, although it's noted that this approach can quickly use up 'fast hours', which is a measure of the speed at which images are generated.

πŸ’‘Fast Hours

'Fast Hours' is a term used to denote a period of accelerated image generation on the Midjourney website. Earning free fast hours by rating images is mentioned as a way to enhance the user's experience and efficiency in creating images.

πŸ’‘Repeat Parameter

The 'Repeat Parameter' is used in conjunction with a small chaos value to explore the full potential of a given prompt. It is presented as an effective method to understand the variations that can arise from a single prompt, emphasizing its importance in the exploration of creative possibilities.

πŸ’‘Smart Folders

Smart Folders are a feature on the Midjourney website that allow for fast and efficient organization of prompts. By adding keywords, the system automatically fills the folder with prompts containing those words, streamlining the process of managing and accessing generated content.

πŸ’‘Negative Prompting

Negative prompting is a technique used to exclude certain elements or occurrences from the image generation process. For example, to avoid generating a red rose, one would use 'dash dash no red' at the end of the prompt. This technique is shown to be effective in controlling the final output and achieving desired results.

πŸ’‘Chaos Parameter

The 'Chaos Parameter' is a setting that introduces randomness into the image generation process. It is mentioned in the context of creating unique and vibrant images, suggesting that higher chaos values can lead to more distinctive and less predictable results.

πŸ’‘Style Reference

A 'Style Reference' is a feature that allows users to direct the AI to generate images in a specific artistic style. It is discussed in relation to high chaos values, indicating that this combination can produce striking and unconventional images.

πŸ’‘Seed Number

A 'Seed Number' is a unique identifier used in the image generation process to produce a particular outcome. Locking in a seed number allows users to test the impact of individual words on the final image, providing a way to refine and control the creative output.

Highlights

Use the image weight parameter (D-iw3) to recreate an exact picture.

Blend up to five images at once with the forward SL blend feature on Discord.

Generate images with a specific color by using that color as a character reference.

Incorporate emojis in your prompt for generating random and unique images.

Write prompts with permutations to save time, but be cautious as it may consume your fast hours quickly.

Earn free fast hours by rating images on the Midjourney website; the top 2,000 raters daily receive one GPU fast hour.

Explore the full potential of a prompt using the repeat parameter and a small chaos value.

Use Smart folders for efficient organization of prompts based on keywords.

Start with a basic idea and then elaborate to refine the generated concept.

In-paint any character in a scene using the very region feature.

Experiment with the 'Memphis pattern' for unique prompts.

Utilize the 'nii' model for anime-style images by appending D-nii6 to your prompt.

Negative prompting can help exclude common elements, such as colors, from your image generation.

Remixing with strong variation can produce fun and consistent art by changing the subject while maintaining the prompt.

Use the 'candid' word in your prompt for a more relaxed aesthetic.

Textures can be used as character references, such as 'Mossy medieval armor' or 'Granite armor'.

Utilize the 'prefer' suffix command in Discord for a preferred aspect ratio in your creations.

Generate images in a thin aspect ratio for bookmark-sized art.

The aspect ratio is crucial for logo creation; consider using a square ratio.

Character concepts can benefit from a vertical poster ratio, such as 2x3 or 4x5.

Fix issues like messed up hands using the subtle variation button or vary region.

Create larger seamless patterns by inpainting a tiled prompt and excluding the outside edges.

Use the chaos parameter for unique image generation, such as a 'vibrant retro pop art picture of a rock and roll penguin'.

Achieve the most realistic images with style raw and a low stylization value.

Higher stylized values can produce the most beautiful images.

Combine stylize, chaos, and weird parameters for the most fun images.

When Midjourney is not responding as desired, use multi-prompting to clarify your request.

Lock in a seed number to test the impact of individual words on the generated image.

Create an outline in Photoshop or Canva, then use it as an image prompt for a rough idea realization.

Check public Midjourney channels with forward SL blend for super user club access.