Midjourney announced the alpha launch of Midjourney v7 on April 4, 2025, aiming to compete with ChatGPT’s recent image generator. The update promises enhanced image quality, AI model customization, and improved user prompt understanding, with weekly updates planned. Meanwhile, ChatGPT’s integration of its image generator, inspired by Studio Ghibli, has attracted significant attention, creating pressure on Midjourney to deliver impactful results. User feedback on text rendering in v7 is being considered for future improvements.
In a bid to reclaim the spotlight from ChatGPT, which has been generating significant buzz lately due to its innovative image generator, Midjourney made a bold announcement on April 4, 2025. Their statement revealed the launch of Midjourney v7, a fresh update that aims to impress the AI art community.
This latest release is an alpha version of the new image-generating model powered by generative artificial intelligence, as the fully polished version is still under development. The alpha phase of Midjourney v7 will be available for an extended period, with regular updates rolling out weekly to enhance its capabilities.
“We are excited to announce that we are entering the alpha testing phase of our new image model v7. This is our most advanced, visually striking, and coherent model to date,” stated Midjourney. “We invite you to explore it and anticipate updates every week or two in the coming months.”
Key Features of Midjourney v7
Many enthusiasts believe it was overdue for Midjourney v7 to make its debut. The previous major model, v6 (in alpha), was released back in December 2023. Even though an intermediate update, Midjourney v6.1, emerged in the summer of 2024, there had been a noticeable lack of major advancements in the months that followed—a significant span in the fast-evolving landscape of Generative AI.
In their blog post, Midjourney outlined the exciting features users can expect from v7. These enhancements include superior image quality, automatic activation of AI model customization, a new “Draft” mode to refine concepts, and two operational modes (relax and turbo) for varied uses.
Additions to v7 will also bring improvements in editing, upscaling, re-texturing, character and object reference modules, and a host of other features. Crucially, Midjourney is focusing on a deeper understanding of user prompts, which is vital for accurately interpreting instructions.
The Competitive Landscape: ChatGPT’s Bold Move
Midjourney is under pressure to deliver impactful results quickly, especially since v7 has been in development for quite some time—hints were dropped as early as 2023. Historically, Midjourney was a leader in AI-driven image creation, seemingly unfazed by OpenAI’s presence in the same arena with DALL-E.
Following its release at the end of March, OpenAI’s latest image generator is now integrated into ChatGPT, the company’s flagship product. This shift suggests a gradual diminishment of DALL-E as an independent tool, as the generator is available to all users, even those without a subscription (though with certain usage limitations).
With ChatGPT’s remarkable popularity boosting visibility for the new generator, OpenAI cleverly incorporated a style reminiscent of Studio Ghibli, renowned for its exceptional film quality (such as My Neighbor Totoro, Porco Rosso, Spirited Away, and Princess Mononoke).
Although this raises potential intellectual property concerns—particularly regarding the training data, as the stylistic similarities hint at influences from Ghibli’s works—it has also garnered extensive media coverage. ChatGPT’s user base has surged beyond what OpenAI initially expected.
As comparisons between Midjourney v7 and ChatGPT’s image generator emerge, not all feedback has favored Midjourney. Users have pointed out that the alpha version’s text rendering could be improved. In response, Midjourney acknowledged these concerns, noting that future development would be guided by community feedback and voting.