TOP Midjourney Prompt Tips, Codes & Tricks

Wade McMaster - Creator Impact
8 Dec 202214:38

TLDRThis video offers a variety of tips and tricks for enhancing AI-generated art using Midjourney platform. It covers prompt codes like aspect ratio control, creative boosts, and algorithm choices between test and testP for different artistic outcomes. Viewers learn how to stylize images, use negative keywords for exclusion, create videos of image creation, and adjust quality settings. The video also introduces custom codes, chaos control, and seed usage for similar image results. It concludes with advice on backing up creations and combining AI with Photoshop for ultimate creative freedom.

Takeaways

  • 🖼️ Use prompt codes like '--AR' to control the aspect ratio of your AI-generated images.
  • 🎨 Add '--creative' to your prompt for more imaginative outputs with the test or test P algorithm.
  • 📸 Experiment with different algorithms by using '--test' for artistic images or '--testP' for more lifelike photographs.
  • 🌟 Utilize '--stylize' to enhance the style of your images, with values ranging from 0 to 60,000.
  • 🚫 Employ negative keywords with '--no' to exclude certain elements from your image prompts.
  • 🎥 Use '--video' to create a time-lapse video of your image's creation process.
  • 🔍 Adjust the quality of your images with the '--quality' command, with higher numbers resulting in more detailed results.
  • 🔄 Explore different versions of the mid-journey algorithm by using '--v' followed by the version number.
  • 🔄 Remix existing images into new creations with the 'remix mode' feature.
  • 🌱 Discover similar results with '--seed' by copying the seed number from a reaction and using it in your prompt.
  • 🔄 Control the unpredictability of your images with '--chaos', adjusting the level of chaos in the algorithm.
  • 🛠️ Set custom codes using '/settings' to streamline your preferred settings for future prompts.
  • 📏 Use word weight to prioritize certain elements in your prompts for more control over the final image.
  • 💾 Backup your images regularly to avoid losing your creations if you cancel your account.

Q & A

  • What is the purpose of the aspect ratio (AR) code in mid-journey AI art prompts?

    -The aspect ratio (AR) code allows users to control the width-to-height ratio of the generated image. By adding '--AR width height' to the prompt, users can specify the dimensions they want, such as making an image twice as wide as it is high.

  • How does the 'creative' tag enhance the AI-generated images in mid-journey?

    -The 'creative' tag, used with the '--test' or '--testP' algorithms, encourages the AI to produce more imaginative and unique images. It helps the AI to go beyond a standard interpretation of the prompt, resulting in more creative and visually interesting outputs.

  • What is the difference between using '--test' and '--testP' in mid-journey AI art prompts?

    -The '--test' algorithm creates artistic-looking images, which are a step up from version three. In contrast, '--testP' uses the photographic algorithm, resulting in more lifelike and realistic images.

  • Can you explain the function of the '--stylize' command in mid-journey AI art?

    -The '--stylize' command lets the AI stylize the image prompts, with a range from 0 to 60,000. The default is 625. By adjusting this value, users can control the level of stylization, influencing how creatively the AI interprets the prompt.

  • How do negative keywords, denoted by '--no', work in mid-journey AI art prompts?

    -Negative keywords, specified with '--no keyword', allow users to exclude certain elements from the generated images. The AI will attempt to create an image that does not include the specified words, helping to refine the results according to the user's preferences.

  • What is the process for creating a video of the image creation in mid-journey?

    -To create a video of the image creation, users add '--video' to their prompt. After the image is generated, they need to add a reaction to the prompt and select the envelope icon to generate the video, which can then be viewed or downloaded as an MP4 file.

  • How does the '--quality' command affect the rendering time and detail of mid-journey AI art?

    -The '--quality' command, with a scale from one to five, determines the rendering time and level of detail in the image. A higher quality setting, such as five, will take longer to render but will produce a more detailed and refined image.

  • What do the '--V' and '--V4' commands do in mid-journey AI art prompts?

    -The '--V' command followed by a number selects the version of the mid-journey algorithm used to generate the image. '--V4', for example, selects the fourth version of the algorithm, which may produce higher quality results compared to the default version three.

  • How can users create custom codes in mid-journey for frequently used options?

    -Users can create custom codes by going to '/prefer option set' and defining a code and its corresponding options. Once set, they can simply type their prompt followed by the custom code to apply the predefined options, streamlining the process.

  • What is the purpose of the '--seed' command in mid-journey AI art prompts?

    -The '--seed' command allows users to replicate similar results by using the same randomized number that initiated the image creation process. By finding and copying the seed number from a previous image and using it in a new prompt with '--seed number', users can generate similar layouts and compositions.

  • How does the '--chaos' command influence the unpredictability of AI-generated images?

    -The '--chaos' command, with a value ranging from 0 to 100, controls the level of chaos and unpredictability in the AI's interpretation of the prompt. A higher chaos value results in less predictable and more varied images.

  • What is the benefit of using the 'remaster' feature with different mid-journey algorithms?

    -Using the 'remaster' feature with different algorithms allows users to take an existing image and reinterpret it using a different algorithm version. This can lead to unique and creative outcomes that might not be achieved with a straightforward render using the same algorithm.

  • How can users ensure they don't lose their mid-journey images?

    -Users can back up their images by logging into their mid-journey account, selecting the images they want to save, and using the downloader feature. They can also archive entire days of images and download them all at once to ensure they have a local copy.

  • Why is it recommended to learn Adobe Photoshop in conjunction with using AI art tools?

    -Learning Adobe Photoshop provides users with more freedom and control over their AI-generated images. They can edit, combine, and manipulate the images to suit their needs, offering a greater range of creative possibilities and ensuring the final result aligns with their vision.

Outlines

00:00

🎨 Enhancing AI Art Creativity with Prompt Codes

The video provides various tips and tricks for enhancing AI-generated art. It starts by introducing prompt codes that can be appended to the end of a prompt to influence the outcome. The first code discussed is for aspect ratio control, which allows users to specify the width and height of the image. The video then compares the effects of using the 'creative' tag with the 'test' or 'test P' algorithms, demonstrating how it can lead to more imaginative results. The 'test' algorithm is shown to produce artistic images, while 'test P' generates more lifelike photographs. The 'stylize' command is introduced next, which can intensify the creative interpretation of the image, with examples provided to illustrate the differences at varying levels of stylization. The video also covers negative keywords to exclude certain elements from the image and the 'video' command to create a time-lapse of the image creation process.

05:01

🔍 Exploring Advanced Features in Mid-Journey AI Art

This paragraph delves into advanced features of the AI art platform. It begins by discussing the 'quality' command, which allows users to adjust the level of detail in the image, with a higher setting resulting in more detailed and time-consuming renders. The video then explores the different versions of the Mid-Journey algorithm, showing how using version 4 can improve the quality of the results compared to the default version 3. The concept of 'remix mode' is introduced, which enables users to create variations of an image by altering the prompt while retaining much of the original composition. The 'seed' command is also explained, which uses a specific number to generate similar results, allowing for the replication of a particular style or composition. Lastly, the 'chaos' command is introduced, which controls the unpredictability of the AI's interpretation of the prompt, with examples showing the stark differences between low and high chaos levels.

10:03

🛠 Customizing AI Art with Personalized Settings and Tools

The final paragraph focuses on customizing the AI art experience through personalized settings and additional tools. It starts by explaining how to set custom codes for frequently used commands, streamlining the process of inputting prompts. The paragraph then discusses the 'suffix' feature, which automatically appends specific settings to the end of every prompt, useful for maintaining consistent output. The video also touches on the use of 'word weight' to prioritize certain elements within a prompt, giving examples of how adjusting the weight can change the focus of the generated image. The importance of backing up images to prevent loss is highlighted, with instructions provided on how to download images from the platform. The video concludes with a recommendation to learn Adobe Photoshop for further control and manipulation of AI-generated images, and to explore other tools for enhancing AI art.

Mindmap

Keywords

💡Aspect Ratio

Aspect ratio refers to the proportional relationship between the width and height of an image or screen, commonly expressed as two numbers separated by a colon (e.g., 16:9). In the context of the video, controlling the aspect ratio is crucial for creating images that fit specific dimensions. The script mentions using '--AR' followed by numbers to set the width and height, as seen in the example where the image is twice as wide as it is high, achieving a 2:1 aspect ratio.

💡Creative Tag

The creative tag is a prompt modifier used in AI art generation to enhance the algorithm's creativity. It is applied to the 'test' or 'test P' algorithm in mid-journey AI to produce more imaginative and unique results. The video illustrates this by comparing a standard turtle image with one generated using the creative tag, resulting in a more stylized and imaginative depiction with added elements like color and goggles.

💡Test Algorithm

The test algorithm is a specific version of the AI's image generation process that can be invoked using the '--test' or '--testP' codes in the script. It is used to produce artistic images, with 'test' yielding artistic results and 'test P' producing more lifelike, photographic images. The script demonstrates the difference by showing the outputs of both algorithms with the same prompt.

💡Stylize

Stylize is a command that allows the AI to apply a certain style intensity to the generated images. It ranges from 0 to 60,000, with 625 being the default value. The video explains how increasing the stylize value to 60,000 results in a more abstract and unique interpretation of the prompt, while reducing it to 20,000 retains some of the original features but still offers a creative take on the image.

💡Negative Keywords

Negative keywords, indicated by '--no', are used to exclude certain elements from the AI-generated images. The script provides an example where an outdoor scene is created with less water and fewer trees by specifying these elements as negative keywords. This helps the AI to focus on the desired aspects of the prompt while minimizing or eliminating undesired elements.

💡Video

The 'video' command in the script is used to create a time-lapse video of the image generation process. It is mentioned that this feature does not work on upscales and requires an additional step to trigger the video creation through the platform's interface. The video script describes how to access and download the resulting MP4 file for personal use.

💡Quality

Quality, denoted by '-q' followed by a number from one to five, is a setting that controls the rendering time and detail of the generated images. The higher the number, the more detailed and time-consuming the rendering process. The script shows a comparison between a default setting and a quality setting of five, highlighting the increased detail and information in the higher quality image.

💡Algorithm Version

The algorithm version refers to different iterations of the AI's image generation process. The script discusses using '-v' followed by a number to select a specific version, such as version three or four. Each version produces different results, with newer versions typically offering higher quality or different artistic interpretations. The script advises checking the settings for the latest available algorithms.

💡Remix Mode

Remix mode is a feature that allows users to create variations of existing images by altering the prompt while retaining the original image's composition and style. The script demonstrates this by changing a robot image to a turtle image, showing how the AI can generate a new image with a similar layout and design but with different content based on the updated prompt.

💡Seed

The seed is a randomized number used by the AI to initiate the image generation process. By using the same seed with different prompts, users can achieve similar results with minor variations. The script explains how to find the seed from an image's reaction options and use it with a new prompt to create a comparable outcome, as shown with the Terminator face example.

💡Chaos

Chaos, controlled by the '--chaos' command followed by a number, influences the unpredictability and randomness in the AI's interpretation of the prompt. The script contrasts an image generated with zero chaos, which closely follows the prompt, with one created with a chaos level of 100, resulting in a less predictable and more varied outcome. Adjusting the chaos level allows users to explore different creative possibilities.

💡Custom Codes

Custom codes are user-defined shortcuts for frequently used combinations of settings. The script explains how to set a custom code through the platform's interface, which can include aspect ratio, video, and quality settings. Once defined, these codes can be quickly applied to new prompts, streamlining the image generation process and saving time.

💡Word Weight

Word weight allows users to prioritize certain words within a prompt, influencing the AI's focus on specific elements. The script uses a prompt with a dog, cat, and mouse to illustrate how assigning different weights to each word can change the prominence of each animal in the resulting image. By increasing the weight of 'mouse', the AI generates an image where the mouse is the main focus, demonstrating the control over the AI's interpretation.

💡Backup

Backup refers to the process of saving and archiving images generated by the AI to prevent loss, especially if an account is canceled. The script provides instructions on how to select and download images from the platform, ensuring that users retain access to their creations. This step is crucial for preserving the results of the AI's work for future use or reference.

💡Adobe Photoshop

Adobe Photoshop is a widely used software for image editing and manipulation. The script suggests learning Photoshop to have more control over AI-generated images, allowing users to edit, combine, and refine their creations. Photoshop skills can significantly enhance the capabilities of AI art by providing users with the tools to achieve their desired outcomes.

Highlights

Introduction to prompt codes for enhancing AI art creativity.

Using aspect ratio code (--AR) to control image dimensions.

Comparing the effects of using the creative tag with and without it.

Differences between test and testP algorithms for artistic and photographic results.

Stylizing image prompts with the --stylize or --s command.

Adjusting stylization intensity with values from 0 to 60,000.

Utilizing negative keywords (--no) to exclude elements from prompts.

Creating a video of image creation with the --video command.

Quality command (--q) to adjust rendering time and detail.

Algorithm version command (--v) to select different rendering versions.

Remix mode for creating variations of existing images.

Using --seed for consistent results in image generation.

Chaos command (--chaos) to introduce unpredictability in image interpretation.

Remastering old algorithm images for unique results.

Creating custom codes for frequently used options.

Using suffix preferences for automatic addition of commands.

Word weight technique to prioritize elements in prompts.

Recommendation to backup images to prevent loss.

Advice on learning Adobe Photoshop for greater image control.

Introduction to additional tools for improving AI-generated images.

Casual Browsing

Midjourney v5 - Style Prompt Tips and Reference Tricks

2024-05-23 10:30:01

How to get started with AI Art using MidJourney - Prompt engineering and tips & tricks

2024-05-22 08:45:03

Midjourney V6 Decoded: Tips, Tricks, and Expert Insights

2024-08-01 07:13:00

Weird Midjourney Tricks with just One Prompt

2024-05-23 04:15:02

Midjourney PROMPT TIPS for more precise control.

2024-05-23 02:50:02