Midjourney Unveils V7: The Exciting New AI Image Model After a Yearlong Hiatus!
Midjourney, a pioneering AI image-generating service, has recently launched its latest model, V7, marking its first major update in nearly a year. This new AI image model has generated excitement following OpenAI’s recent viral image generator release in ChatGPT, known for producing enchanting Ghibli-style illustrations. However, V7 is not specifically designed for Ghibli aesthetics, yet it still promises to deliver stunning visual outputs.
Overview of the New Midjourney V7 Model
The V7 model began its alpha rollout around midnight EST on Thursday, just a week after OpenAI’s announcement. Midjourney’s CEO, David Holz, describes V7 as “a totally different architecture” that significantly enhances the user experience. Here’s what you need to know:
- Enhanced Text Prompt Understanding: V7 is designed to better comprehend and respond to text prompts, delivering higher quality images with improved textures.
- Improved Coherence: The model shows noticeable advancements in rendering bodies, hands, and various objects with greater detail.
- Two Modes Available: Users can choose between Turbo (higher cost) and Relax modes, along with the new Draft Mode that accelerates image rendering.
Getting Started with V7
To utilize the V7 model, users must first create a personalized profile by rating approximately 200 images. This profile tailors the AI’s outputs to match your visual preferences. Importantly, V7 is the first model that has personalization activated by default.
Once your profile is set up, you can easily enable or disable V7 through Midjourney’s website or its Discord chatbot. In the web application, simply select the model from the drop-down menu next to the “version” label.
Key Features of Midjourney V7
Here are some of the standout features of the V7 model:
- Draft Mode: This feature allows for images to be rendered at 10 times the speed and half the cost compared to the standard mode, although the quality is lower.
- Future Updates: While some functionalities like image upscaling and retexturing are not yet available, Holz mentioned that these features should arrive within the next two months.
- User Feedback: Midjourney is eager to receive user input to refine the model’s capabilities, encouraging experimentation with different prompting styles.
Midjourney’s Unique Position in the AI Landscape
Founded in 2022 by Holz, who previously co-founded Leap Motion, Midjourney has not taken any outside funding. The company is reportedly on track to generate around $200 million in revenue by late 2023. Recently, Midjourney announced the formation of a hardware team to explore new projects while continuing to enhance its AI models for video and 3D object generation.
However, Midjourney is currently facing several lawsuits alleging copyright infringement for using images sourced from the web without the consent of the original creators.
For more information about Midjourney and to stay updated on the latest features, visit their official website or follow their Twitter account.
