Midjourney V6 vs V7: Side-by-Side Visual Differences

midjourney blog post image

Testing Midjourney Versions

Comparing different versions of Midjourney can be insightful. One way to do this is by using the same prompts and running them through both versions. Then, you can look closely at the resulting images.

To generate prompts for this test, several large language models (LLMs) were used, including Deepseek, Claude, Perplexity, and Grock. Each LLM was given a concept, and asked to create a Midjourney prompt. Then, each prompt was used in both Midjourney v6 and v7.

A Specific Example: The “Hands” Concept

The concept used for this comparison was “a bunch of hands.” The prompts generated by the LLMs based on this concept were simple. These same prompts were used for both Midjourney v6 and Midjourney v7.

Noticing the Visual Shifts

When comparing the images side-by-side, a clear visual difference appeared:

  • The images generated with Midjourney v6 tended to be darker and had a moodier feel.
  • It was sometimes harder to clearly see the details in the v6 images because of the lower light levels and presence of shadows.
  • The images from Midjourney v7 were notably lighter.
  • Many of the v7 results showed scenes with white sand and often featured hands belonging to people of color, with less prominent shadows.
  • The details, like the hands themselves and the texture of the sand, were easier to distinguish in the v7 images.

This difference was consistent across the images generated from the various LLM prompts. The version six outputs were generally darker and obscured, while the version seven outputs were consistently lighter and clearer, frequently depicting white sand backgrounds and specific skin tones.

Same Prompt, Different Look

It’s interesting because the prompt itself was identical for each pair of images (v6 and v7 outputs from the same prompt). This shows that the underlying model changes between Midjourney versions can lead to different artistic interpretations and visual styles, even when given the exact same instructions.

Understanding the Evolution

While the specific technical details of how Midjourney’s models change from v6 to v7 aren’t publicly known, visually comparing the output reveals an evolution in the model’s style. Version 7 seems to have shifted towards a lighter palette and potentially different interpretations of prompt elements compared to version 6.

Conclusion

This simple comparison of identical prompts in Midjourney v6 and v7 highlights that updating versions can result in a noticeable change in the visual style of generated images. Version 6 showed darker results, while Version 7 trended towards lighter scenes, often with specific elements like white sand and clearer details. Watching for these shifts helps us understand how the AI is evolving and what to expect when creating images.

🖥

Do This Locally with Logen

Skip the subscription. Generate images like these on your own GPU — private, unlimited, no cloud dependency.

Try Logen Free →

Midjourney vs Logen

FeatureMidjourneyLogen
Cost$10-60/moFree & Pro plans
PrivacyCloud (images stored)100% local
GenerationsLimited by planUnlimited
Internet RequiredAlwaysNo
Custom ModelsNoYes (any ComfyUI model)
GPU RequiredNo (cloud)Yes (local)
Ease of UseDiscord botDesktop app
Try Logen Free →
FREE 11-PAGE PDF HANDBOOK + LOCAL AI FIELD NOTES

Get the Free 11-Page Local AI Image Generation Handbook

See what GPU VRAM you need for FLUX.1-schnell and SD 1.5, how checkpoints and LoRAs work, and five repeatable workflows on Windows and Linux x64. Enter your email to open the 11-page PDF right away and get our six-part weekly local AI field notes.

6–24 GB VRAM Tables Cloud vs. Local Math 5 Studio Workflows
How signup uses your email

We register your email with the TitanXT newsletter on beehiiv to send the 11-page PDF link, our six-part weekly local AI sequence, and occasional Logen updates. You can unsubscribe in one click from any email or by writing to support@titanxt.io. Read beehiiv's privacy policy (new tab).

Scroll to Top