Midjourney V8.1 becomes the default model, delivering 4x resolution and 4-second image generation
Midjourney set V8.1 as its default model on June 11, 2026, replacing V7 after user testing and feedback. The transition marks V8.1's general availability as the standard experience for all Midjourney users, bringing significant speed and resolution improvements to the platform's default output.
What's new
With V8.1 now default, every Midjourney user receives V8.1 outputs without any configuration change. Key improvements over the previous default (V7):
- Speed — Standard definition images generate in approximately 4 seconds; HD images in 12 seconds
- Resolution — With HD mode enabled, V8.1 renders at twice the size and 4x the resolution of V7 images, reaching effective output resolution in the 2K range
- Prompt adherence — The model is described as "smarter, more coherent, better adheres to detailed prompts, and renders text better than ever" compared to V7
- Style continuity — Style references, personalization settings, and aesthetic preferences remain consistent between V7 and V8.1, making transitions backward-compatible for users with established workflows
Midjourney is deprecating V8.0 alpha within two weeks of the June 11 announcement. The V7 omni-reference feature remains available as V8.1's equivalent is still being developed.
Context
V8.1 first appeared as an alpha in early 2026, with Midjourney running extensive blind rating sessions to validate quality before setting it as default. The company temporarily set Standard Definition as default during a server capacity transition period, then restored HD as the standard as infrastructure stabilized.
The transition from V7 to V8.1 continues Midjourney's pattern of overlapping generation cycles: keeping legacy features (like V7 omni-reference) accessible while the new default matures. This approach reduces churn among professional users who have built production workflows around specific model behaviors.
Midjourney has not publicized architectural details about V8.1's training or parameter count, consistent with its longstanding practice of limiting technical disclosure. The company remains privately held and does not publish model cards or technical reports in the manner of Google, Anthropic, or Meta.
Why it matters
Image generation speed is a direct usability multiplier: the difference between a 4-second and a 25-second generation cycle changes how designers and creative professionals use the tool in practice. At 4 seconds, image generation approaches interactive speed — faster than a typical save-and-reload cycle in most design software.
The 4x resolution improvement in HD mode addresses a persistent complaint about AI image generators: that default output quality is insufficiently sharp for print or high-fidelity display applications. Rendering at near-2K resolution by default moves Midjourney closer to production-ready output for commercial clients.
Midjourney remains one of the most widely used image generation platforms among creative professionals, despite competition from Stable Diffusion derivatives, Adobe Firefly, DALL-E, and newer entrants like Ideogram and Black Forest Labs' Flux. V8.1's speed and resolution gains reinforce the platform's competitive positioning as Google, OpenAI, and others push into the image generation market through their respective ecosystems.
Corroborating sources
- Updates.midjourney
https://updates.midjourney.com/v8-1-is-now-the-default-model/
“The model is smarter, more coherent, better adheres to detailed prompts, and renders text better than ever”