6.1? | Midjourney Office Hours Recap Feb 7th 2024 | Midjourney News

Future Tech Pilot
8 Feb 202404:12

TLDRIn the Midjourney Office Hours Recap from February 7th, 2024, a new feedback system is announced for the following week, aiming to improve user satisfaction. The website's version six is set to become the default, enhancing performance in prompts, coherence, and textures. Upcoming features include social aspects and an improved creation flow. The team is also focusing on character consistency and a new 'describe' version for better accuracy. Patreon adjustments offer multiple prompts for a $5 subscription. The script hints at a potential version 6.1 with a new style reference system and personalization features, alongside faster 'turbo mode' for V6. Future work includes control net and drawing input features, with version 7 promising further image quality improvements.

Takeaways

  • 🆕 A new feedback system will be introduced next week to improve understanding of user desires.
  • 🛠️ Version 6 of the website is being refined for better performance in prompts, coherence, and textures, and will soon become the default.
  • 📈 Work is underway on the next version of the website with a focus on adding more social features, although specifics are still under wraps.
  • 📅 Expect design polishing, onboarding, and educational features to be prioritized closer to March.
  • 🔧 A new system for feedback and bug reporting is in development to streamline user communication.
  • 🔄 An updated 'describe' feature is in the works for improved accuracy, with a hopeful release within a week.
  • 🎭 Significant efforts are being dedicated to enhancing character consistency, with promising results expected in a few weeks.
  • 🎨 Patreon adjustments offer multiple prompts per week for a $5 subscription, including access to 36 existing prompts.
  • 🔄 A new style reference system is being developed to merge with the character system, potentially offering a checklist for consistency in images.
  • 🔄 Version 6.1 is anticipated to include a multi-dimensional referencing system for model focus, with personalization features to follow.
  • 🚀 A 'turbo mode' for V6 is being prototyped, promising faster generation times at a higher cost, suitable for group use.
  • 🎨 Features like control net and drawing input are in the experimental phase, with outcomes yet to be determined.
  • 🔮 Version 7 is on the horizon with expected upgrades in image quality, coherence, knowledge, and prompt understanding, but it's still several months away.
  • 🎥 The video feature is taking longer than expected, with no imminent release, but early tests show promise.

Q & A

  • What is the main topic of the Midjourney Office Hours Recap for February 7th, 2024?

    -The main topic is the upcoming updates to the Midjourney platform, including a new feedback system, improvements to the creation flow, and the introduction of social features in the next version of the website.

  • What improvements are expected in the new feedback system mentioned in the recap?

    -The new feedback system is expected to provide a better understanding of what users actually want, although specific improvements are not detailed in the recap.

  • What is the significance of version six being switched to the default soon?

    -Switching to version six as the default indicates that users can expect better performance for prompts, coherence, and textures in their creations.

  • What are the social features that the next version of the Midjourney website will include?

    -The exact social features are not specified, but the mention of 'dramatic change' suggests significant enhancements to the user interaction and community aspects of the platform.

  • What does the speaker mean by 'character consistency' in the context of Midjourney updates?

    -Character consistency refers to the platform's efforts to ensure that generated characters remain consistent across different images, which is a feature they are actively working on improving.

  • What is the purpose of the new version of 'describe' that the team is working on?

    -The new version of 'describe' aims to be more accurate in understanding and generating descriptions, although the exact improvements are not specified in the recap.

  • What adjustments were made to the speaker's Patreon, and what do the supporters receive for $5?

    -For a $5 contribution, supporters gain access to multiple prompts per week, with 36 prompts already posted, offering a good value for their support.

  • How does the new style reference system being developed relate to character consistency?

    -The new style reference system is expected to merge with the character system, potentially providing a checklist for users to maintain consistency in aspects like outfits or character features across images.

  • What is the 'turbo mode' mentioned for version 6, and how does it differ from the standard version?

    -The 'turbo mode' is a faster version of V6, which may be three times quicker but more expensive. It is designed for generating images with groups or larger batches.

  • What are the speaker's thoughts on the potential 'control net' and 'drawing input' features?

    -The speaker is unsure about the effectiveness of these features but mentions that the team will train the models to see how they perform.

  • What can users expect in version 7 of Midjourney, according to the recap?

    -Version 7 is expected to bring the usual upgrades such as better image quality, coherence, knowledge, prompt understanding, and pixel image quality, with improvements to smaller details in the images.

  • What is the current status of video generation capabilities in Midjourney, and what can users expect?

    -Video generation is taking longer than anticipated, and users should not expect it soon. However, when it becomes available, it is expected to offer good results.

  • What are some of the user requests that the Midjourney team is aware of and working on?

    -The team is aware of and working on user requests for more style control, camera control, pose control, and better human anatomy in the generated images.

Outlines

00:00

🚀 Mid Journey Office Hours Recap & Upcoming Features

The script provides a detailed recap of the Mid Journey office hours from February 7th, highlighting the introduction of a new and improved feedback system, set to launch the following week. The main focus of the team is to refine the creation flow on the website, with version six expected to become the default soon, promising better performance in prompts, coherence, and textures. The next version of the website is also in the works, featuring more social features, although specifics are yet to be determined. The script mentions the need for a new system for feedback and bug reporting, as well as ongoing development of a more accurate version of 'describe'. Character consistency is a key area of focus, with promising developments expected to be finalized in a few weeks. The Patreon is also mentioned, offering multiple prompts per week for a $5 subscription, and adjustments have been made to enhance the value. The script concludes with a tease of version 6.1, which may include a new referencing system with multi-dimensionality, and a personalization feature that is still in the early stages of development.

Mindmap

Keywords

💡Version 6

Version 6 refers to the sixth iteration of the Midjourney platform's AI model, which will soon become the default version. It promises improvements in prompt coherence, texture quality, and overall performance. This upgrade is significant as it sets the foundation for future enhancements, including new features and better user experiences.

💡Feedback System

The new feedback system is an upcoming feature in Midjourney, aimed at better understanding user preferences and improving the platform's functionality. It's expected to launch soon and will offer a more efficient way for users to communicate their needs and issues, ultimately enhancing the development process.

💡Social Features

Social features are part of the planned upgrades for the Midjourney website. These features will allow users to interact more closely, potentially sharing creations and collaborating. This addition is expected to significantly change the user experience, making the platform more community-oriented.

💡Character Consistency

Character consistency refers to the AI's ability to maintain uniformity in characters' appearance and attributes across different images. This is a crucial aspect that users have highlighted as a priority, and improvements in this area are being actively developed. It involves ensuring that details like outfits and features remain consistent.

💡Describe Feature

The 'describe' feature is being refined to provide more accurate descriptions of images generated by the AI. This tool helps users understand and articulate the content and style of the images created, making it easier to replicate or modify results. It's a key part of the user interface improvements planned for Midjourney.

💡Style Reference System

The style reference system allows users to specify stylistic elements they want the AI to follow in generated images. An updated version of this system is being developed to work seamlessly with character consistency, potentially offering users a checklist to maintain or alter specific image aspects, like outfits or poses.

💡Turbo Mode

Turbo mode is a proposed feature that would significantly increase the speed of image generation. Although it would come at a higher cost, it offers a valuable option for users who need faster output, especially useful for generating large batches or working with groups.

💡Version 7

Version 7 is the next major update planned for the Midjourney platform, expected to include improvements in image quality, coherence, prompt understanding, and more detailed visual elements. It's still several months away, and its development includes addressing user desires for more control over style, camera angles, and human anatomy.

💡Control Net

Control Net refers to a potential feature that could introduce more control over image elements, possibly allowing for drawing inputs or more precise control over the output. This concept is still in an exploratory phase, with uncertainty about its effectiveness and implementation.

💡Video Generation

Video generation is an ambitious goal for Midjourney, focusing on extending their AI capabilities to create video content. While it's still in the early stages and facing delays, the feature promises to offer dynamic and high-quality video creation once it's fully developed.

Highlights

A new feedback system is coming next week, aiming to provide better understanding of user needs.

Version six of the website will be switched to the default soon, promising better performance in prompts, coherence, and textures.

The next version of the website will include more social features, although the specifics are not yet clear.

After adding social and creation features, the website will undergo design polishing and onboarding and educational feature enhancements.

A new system for feedback and bug reporting is in development, expected to be needed closer to March.

A more accurate version of 'describe' is under development, with the developer expressing satisfaction with its progress.

Character consistency improvements are in the works, with promising results expected to take a few weeks to perfect.

Version 6.1 is anticipated to include a new style reference system that may offer a checklist for image consistency.

Personalization features are being considered for future updates, although the current progress is behind schedule.

A new tuner is planned, focusing on personalizing the model's biases rather than style.

Faster versions of V6, referred to as 'turbo mode,' are being prototyped, offering speed at a higher cost.

Control net and drawing input features are being explored, with uncertain outcomes.

Version 7 is in the early stages of development, with expected upgrades in image quality and prompt understanding.

Video capabilities are in development but are taking longer than anticipated, with no immediate release expected.

User feedback has highlighted the desire for more style control, camera control, pose control, and improved human anatomy.

The art prompt for the video is provided, showcasing an anime style figure in a colorful graffiti style.

The creator gives a special thanks to a community member for their prompt and invites viewers to engage on Instagram.

Patreon support is offered for exclusive prompts, with an existing collection of 36 prompts available.