Is Midjourney 7.0 coming in 2024? Everything we know so far

Eight months since the release of Midjourney V6 and users are waiting with bated breath for the release of Midjourney 7. It’s anticipated that the latest version will surpass previous iterations with photorealistic rendering, cementing Midjourney as the leading generative AI image model. 

In this article, we’ll discuss some of the speculations surrounding the upcoming Midjourney 7 release and the features that AI artists hope to see introduced. But first, let’s define generative AI and touch on the evolution of this model. 

What is Generative AI? 

Generative AI is all around us–even when we don’t realize it. It’s a type of artificial intelligence technology that quickly creates content, including text, images, videos, etc., based on what it’s asked (known as prompting). While there’ve been massive updates and advancements in recent years, this technology isn’t exactly new.

In the 1960s, generative AI was first introduced with the creation of chatbots. However, it wasn’t until 2014 that generative AI could create convincing audio, video, and images of real people. Through analyzing thousands of artworks and images, AI has learned to generate its own images. This breakthrough brought us the AI image generating tools we use every day.

Common Generative AI Image Models

Over the last few years, leading AI image generators, like Midjourney, Dalle-3, and Stable Diffusion, have taken the world by storm. With an average of 34 million images generated per day, they’ve changed the way many industries operate. As they continue to improve, we can expect to see AI-generated images incorporated into content creation, design, and marketing. 

DALL·E 3 

DALL·E 3 is an incredibly easy-to-use generative AI image model created by OpenAI, the founders of ChatGPT. It’s credited for translating nuanced and detailed prompts into accurate images. Built natively on ChatGPT, it’s currently available through a paid subscription to ChatGPT Plus. ChatGPT serves as your brainstorming partner to help bring your images to life. DALL·E 4 is supposed to come sometimes in late 2024.

Stable Diffusion 

Stable Diffusion is a generative AI image model created by Stability AI. It’s the only major AI image generator that’s open source and provides free credits to users. Stable Diffusion uses DreamStudio, an intuitive app that gives you greater control over your generated images.

The latest Stable Diffusion 3.0 has significant improvements, including better handling of multi-subject prompts, enhanced image quality, and more accurate text generation. The big advantage is Stable Diffusion models is its accessibility and openness.

Flux.1

On August 1, 2024, Black Forest Labs, founded by the creators of Stable Diffusion, released the Flux.1 suite of models. This new open-source generative AI image model creates high-quality images based on natural language text prompts. Images generated by Flux.1 have gone viral online for their shocking realism that’s said to rival that of Midjourney. Another great feature of Flux.1 is it’s capability to generate consistent and readable text within the images.

What is Midjourney?

Founded in 2022 by David Holz, Midjourney is an independent, self-funded research lab that has quickly become one of the biggest names in AI. Their generative AI image model (also called Midjourney) converts natural language prompts into images and is regarded as one of the most popular AI image generators on the market.

Midjourney is accessed entirely through the Discord chat app, which, unfortunately, means your images are public by default. Unlike other generative AI models that offer a few image generations for free, Midjourney has removed its free trial. It offers four subscription tiers, ranging from $10 up to $120 monthly.

Since its initial release, Midjourney’s image generation has improved by leaps and bounds. Several images have gone viral on social media, and one AI-generated image won first place in the digital arts/digitally manipulated photography category at the Colorado State Fair Fine Arts Competition.

¿Quieres desatar tu creatividad en un Mac? ¡Prueba AI Image StudioUsando el último Modelo de Stable Diffusion, genera imágenes fotorrealistas y precisas, dando vida a tus visiones con facilidad. Perfecto para artistas, diseñadores y cualquiera en el medio. ¡Pruébalo y transforma hoy tus ideas en impresionantes visuales!

The Evolution of Midjourney

Today, Midjourney is considered one of the best AI image generators available, with millions of daily users. However, comparing the results of V1 (February 2022) to the quality of images generated with V6 (December 2023), the improvements have been astronomical.

Midjourney 1,2, and 3

Looking back on the images produced with V1, especially the character design images, we might consider them poor quality or wonky. While some questioned whether AI image generation was worth pursuing, others were amazed by Midjourney’s capabilities. 

There isn’t a significant difference from V1 to V3 (July 25, 2022). At that time, images were generated in low detail and needed to be remastered to improve the details and quality. However, Midjourney was committed to improving the prompt understanding and image quality of their product.

Midjourney 4

When V4 was released on November 5, 2022, this was a turning point for Midjourney. The unprecedented improvements in the quality of their images far surpassed their previous iterations, making it the highest-quality model available at the time of release. But their improvements didn’t stop there.

Midjourney 5

Four months after V4, in March 2023, Midjourney released V5. It further enhanced the quality, realism, and versatility introduced in V4 to improve the aesthetics of the images. However, with this version, the flawless finish of the images generated subtracted from the overall realism. 

Midjourney 6

On December 21, 2023, Midjourney released V6 after nine months of development. It significantly improved prompt understanding, image quality, and realism. V6 generates character portraits with more realistic features and asymmetrical structures that more closely resemble real people. 

V6 has also incorporated small details in the foreground and background of landscapes that better emulate real photography. While Midjourney struggled with accurate text generation from V1 to V5, with V6, they became one of the best generative AI image models when it comes to text.

Midjourney 6.1

Midjourney’s V6.1 image model introduces key improvements over previous versions, enhancing image coherence and quality, particularly in rendering complex features like hands, bodies, plants, and animals. The update offers more precise small details, improved texture quality with new 2x upscalers, and approximately 25% faster image processing. Text accuracy in generated images has also been refined.

V6.1 includes a new personalization model with better nuance and accuracy, along with personalization code versioning for continuity in user projects. However, it does not feature a new inpainting model, with those functions reverting to V6.0. Users can switch back to V6 if preferred. Midjourney plans to release V6.2 with further improvements based on user data.

What’s Expected from Midjourney 7?

While official information on what we can expect from Midjourney 7 is scarce, the rumor mill is buzzing with leaked information and speculations that hint at some groundbreaking features for this latest version. Here’s what we can expect from Midjourney V7:

• Realism: With each new version of Midjourney, we’ve seen an increase in realism. It’s safe to assume that V7 will elevate image quality, providing an even higher level of realism. 
• Nuance: Many AI image generators struggle to understand nuance in complex prompts. If Midjourney V7 continues on its expected trajectory, this version is expected to better interpret intricate prompts and generate images that are exactly what the user intended, rivaling that of DALL·E 3. 
• 3D Rendering: Rumors within the community suggest that V7 could include 3D rendering capabilities to enable users to generate files for 3D printing. 
• Character Rendering: Based on the improvements seen in V6, we can expect an improvement in the accurate portrayal of human anatomy, poses, and interactions in V7. 
• Consistency: The last few versions of Midjourney have honed in on consistency. If they follow the same path, we can expect to see better control of style and character reference features. 

Overall, Midjourney appears to be preparing itself to elevate its image quality through much more realistic and consistent designs at a level likely to surpass the competition.

What Users Would Like to See from V7

Although Midjourney is widely popular and typically generates high-quality images, they’re constantly working on improving its program and meeting the needs of the community. Here are some of the improvements the community would like to see:

• Multiple Character Angles: Generating images of the same character or scene from multiple angles while maintaining consistent designs in the output. 
• Consistent Character Reproduction: Generate a character or item from a seed (specific starting point or prompt) and maintain the same appearance across multiple images. This would prevent inconsistencies like a character having different clothes or accessories. The ability to use multiple seeds would ensure several objects stay consistent throughout multiple scenes. 
• Color Palette Control: The ability to dictate and edit the color palette of the generated images. 
• Sketch-to-image: The option to upload a sketch that Midjourney could use as a blueprint to generate a final image. Also, the option of determining how similar the output should be to the reference image while maintaining the stylistic parameters requested. 
• Font Control: Increased control over the font to avoid additional editing after download. 
• Editable Background: Transparent background option to avoid using third-party applications or separate editable layers for the character and background. 

Fortunately for Midjourney users, this company actively listens to the community’s suggestions and complaints. They’ve created a dedicated page where users can submit and vote on ideas for improvements and then tailor its updates to address their priority concerns. 

While there’s no official confirmation yet, we can expect that some of these suggestions will be incorporated into V7. 

When will Midjourney 7 be Released?

Midjourney has been extremely careful about revealing any information on the latest version of its product. While the team has not officially announced anything, the community expects this latest release to be just as much of a surprise as V6 was last year. 

Considering that V6 was released in December 2023 after nine months of development, it’s been speculated that V7 will be released sometime in September 2024 at the earliest. 

However, on June 30, 2024, Miidjourney released V6.1 and announced their plans to release V6.2 in the months that followed. Their official announcement said, “We will likely release a V6.2 model in the next month or so with further improvements, but first, we need to gather as much data as possible of people using V6.1.” 

While estimating the exact timeline for the V7 release is nearly impossible, we can expect that this will be one of Midjourney’s biggest model improvements thus far.