Midjourney has updated how style references work. Most srefs on this website now require to specify version with

--sv 4
Copy
to maintain the same style. Learn more.

--sv 4
Copy
I trained it on Scenario using 27 images initially made on Midjourney with --sref 2314938599 (very grateful to @steelparsley for this code ) Surprisingly... this model did well without ANY captions (I also tried with captioning, with lesser quality outputs)
by
Emm
I trained it on Scenario using 27 images initially made on Midjourney with --sref 2314938599 (very grateful to @steelparsley for this code )

Surprisingly... this model did well without ANY captions (I also tried with captioning, with lesser quality outputs)
--sv 4
Copy
I trained it on Scenario using 27 images initially made on Midjourney with --sref 2314938599 (very grateful to @steelparsley for this code ) Surprisingly... this model did well without ANY captions (I also tried with captioning, with lesser quality outputs)
by
Emm
I trained it on Scenario using 27 images initially made on Midjourney with --sref 2314938599 (very grateful to @steelparsley for this code )

Surprisingly... this model did well without ANY captions (I also tried with captioning, with lesser quality outputs)

Unlock 1020 Premium SREF codes

Reveal all --sref codes from this page curated manually with unique reference styles

Premium sref 1
Premium sref 2
Premium sref 3
Premium sref 4
Premium sref 5
Premium sref 6
Premium sref 7
Premium sref 8
OSZAR »