How to Get Consistent Image Styles with Midjourney’s Sref

🖥

Do This Locally with Logen

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

Try Logen Free →
  • The default value for –sw is 100.
  • You can set –sw from 0 to 1000.
  • Higher –sw values mean the style reference has more impact.
  • Note that the effect might increase a lot from –sw 1 to 50, then level off at higher values.

If your style reference image has a very strong look, it might overpower your text prompt, especially at the default –sw of 100. For instance, applying a photographic style reference (like the bubbles image mentioned earlier) to a prompt for a "children's book illustration" at –sw 100 might result in images that look photographic, not like illustrations.

Lowering the –sw value can help. Reducing it to something like –sw 15 could help Midjourney keep the illustration style from your prompt while still pulling in the color palette from your style reference.

Using Multiple Style References

You aren't limited to just one style reference. You can include multiple image URLs after –sref, separated by spaces.

Example: –sref image1.jpgimage2.jpg

Midjourney will try to blend the styles from all the provided images. This is useful if you want to combine elements from different styles or if adding a second style reference (like an illustration style image) helps steer the generation closer to your desired outcome when a single sref is not enough.

Relative Weights for Multiple Styles

Want one style reference to have more influence than others? You can assign a relative weight using double colons (::) followed by a number after each image URL.

Example: –sref style-image-a.jpg::1 style-image-b.jpg::3

In this example, "style-image-b.jpg" would have three times the influence of "style-image-a.jpg" on the final style.

Saving Style References with Custom Options

Typing out long image URLs repeatedly can be tedious. Midjourney lets you save frequently used parameters, like style references, as custom options.

Setting a Custom Option

Use the /prefer option set command.

Type `/prefer option set`. The first field is for the "name" you want to give your custom option (e.g., `pastel-neon`). The second field is for the "value," where you paste your style reference URL (or multiple URLs and weights). Press Enter.

Now, you have a custom option saved with that name.

Using a Custom Option in a Prompt

In your prompt, instead of typing –sref followed by the long URL, type –sref followed by a dash and the name of your custom option (e.g., –sref –pastel-neon).

Midjourney will recognize the custom option name and use the saved style reference URL(s).

Managing Custom Options

To see all your saved custom options, type /prefer option list.

To delete a custom option, type /prefer option set, enter the name of the option you want to delete, and press Enter without putting anything in the value field.

Sref Works with Niji V6

Good news if you enjoy the anime style: –sref also works with Midjourney's Niji V6 model. Just include –sref along with –niji 6 in your prompt.

Conclusion

Midjourney's –sref parameter is a straightforward way to achieve consistent styles across your image creations. By understanding how to use single or multiple style references, adjusting their strength with –sw, and saving your favorites as custom options, you gain greater control over the final look of your images. Experiment with different reference images and see how they change your results!

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 →
🖥

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