In the event you’ve ever performed with AI picture mills, you most likely know Midjourney. And for those who’ve been round for the reason that early days, then yeah — you’ve seen issues.
I’ve been utilizing Midjourney since model one, and belief me, these early outcomes have been tough. Each immediate felt like rolling the cube: would I get an honest picture or a cursed gremlin with additional enamel and three shadows? You by no means knew.
However that’s what makes this journey so fascinating. Watching an AI go from awkward, clunky experiments to full-blown, professional-grade imagery in only a few updates is wild. And now that V7 is out, I figured it was the proper time to look again and ask: how far have we actually come?
So I ran the identical immediate by each model — from V1 to V7 — and the transformation speaks for itself. Let’s have a look.
Midjourney’s Enchancment Advised Via Photos
Let me take you on a journey by the evolution of AI picture era. I’ve spent numerous hours taking part in with Midjourney’s fashions, from the early days of V1 to the most recent V7 launch, and the transformation is nothing wanting mind-blowing. Utilizing the very same prompts throughout all variations, I am going to present you simply how far we have come.
Portrait (Human)
Immediate: a younger man in a plain black prime, indie, retro, medium format pictures, heat mild, dorm room aesthetics, candid
The early days of Midjourney have been… tough, to place it kindly.
From V1 to V3, you may kinda inform it was trying to generate a person, however the anatomy was all types of improper. Heads have been formed like malformed potatoes, positioned at unimaginable angles, and the dorm background seemed like an summary portray quite than a residing house. It was the uncanny valley, besides the valley was extra like a bottomless pit.
V4 marked the primary main breakthrough, lastly understanding the fundamental form and type of a human being. The anatomical nightmares have been gone, however one thing nonetheless felt off — that traditional uncanny valley feeling that makes you uncomfortable with out figuring out precisely why.
By V5, the technical points have been largely resolved, however as a substitute of an informal candid shot, we obtained one thing that seemed like knowledgeable pictures session. V6 improved issues additional however nonetheless struggled with shadows and lighting.
Then got here V7, and it is like somebody really snapped a photograph with their smartphone. The informal posture, the pure lighting, the genuine dorm setting — it lastly nailed what I used to be asking for all alongside.
Portrait (Non-Human)
Immediate: gray british shorthair cat, medium shot, grainy disposable
If you wish to see nightmare gasoline, simply take a look at what V1 did with a easy cat immediate. The consequence seemed like a tumor fabricated from fur — one thing that might ship kids operating and screaming.
V2 and V3 confirmed some enchancment, however the eyes have been nonetheless disturbingly improper, and there wasn’t a lot evolution between these two variations. Taking a look at these photographs, you’d assume the AI had by no means really seen a cat earlier than.
By V4 and V5, issues obtained considerably higher. The cats really seemed like cats, however there was nonetheless that telltale AI giveaway — fur that blended collectively in an unnatural manner, missing the person strands you’d see in an actual picture.
V6 nailed it with a extra life like fur definition and texture that might cross as an actual grainy disposable digital camera shot. Curiously, in my view, V6 really outperformed V7 for this one. The most recent mannequin’s picture had some unusual lighting points and even some tearing close to the cat’s ear.
Panorama
Immediate: the view from the height of a mountain, sea of clouds, huge and mesmerizing panorama, Pictures, captured with a Fujifilm GFX 100S medium format digital camera
The early variations of Midjourney actually struggled with landscapes. V1 by V3 appeared confused in regards to the elementary distinction between mountains and clouds — they blended collectively in a disorienting mess. The attitude seemed off, scale was all improper, and the entire scene felt like a fever dream quite than an imposing mountain view.
V4 began to grasp the task higher however nonetheless had main points with perspective and scaling. It additionally randomly inserted an individual I by no means requested for within the immediate (a traditional AI hallucination second). By V5, these issues have been largely fastened, with a lot better distinction between mountains and clouds, and a extra pure sense of scale.
The leap to V6 and V7, although? The atmospheric lighting, the feel of the clouds, the dramatic mountain peaks — these newer variations created photographs that might simply cross for skilled panorama pictures. The distinction between V6 and V7 right here was minimal, suggesting that Midjourney is perhaps approaching the ceiling for this specific kind of images.
Product Pictures
Immediate: business pictures, a scented candle, on pastel purple background, with flowers, minimal, dreamy, comfortable lighting, middle composition
Early variations of Midjourney had no thought what a candle ought to seem like. V1 and V2 produced unusual, summary interpretations that hardly resembled something cylindrical, not to mention a useful candle.
V3 lastly grasped the fundamental form of a candle however nonetheless had that apparent “AI-generated” high quality that plagued early text-to-image fashions — flat lighting, unusual textures, and an general lack of realism.
By V4, the mannequin may produce a recognizable candle, however the consequence was painfully plain and lacked any understanding of the way to render flowers or leaves realistically. V5 improved issues however nonetheless did not have that skilled product pictures really feel with the dramatic lighting and distinction that makes a product pop off the web page.
V6 and V7 fully reworked the sport — they each produced photographs that might simply be mistaken for skilled product pictures, with completely centered candles, heat inviting lighting, and background flowers that complemented quite than distracted from the topic.
Pixel Artwork
Immediate: pixel artwork scene, a legendary medieval city with fog, darkish fantasy, 8-bit sport
Pixel artwork continues to be an interesting case examine in Midjourney’s evolution.
V1 by V3 have been objectively horrible, however I used to be shocked that even at this early stage, the AI may acknowledge and generate the fundamental shapes of medieval towers — one thing it struggled with in different picture classes. The outcomes have been crude and barely recognizable, however hey, the basic understanding was there.
Paradoxically, V4 would possibly really be essentially the most profitable model for true pixel artwork. It produced photographs with the genuine chunky pixels and restricted colour palette that outline the style.
As Midjourney superior to V5, V6, and V7, the outcomes turned more and more detailed and “HD” — which is definitely counterintuitive to my immediate. The newer variations appear to simulate pixel artwork quite than create precise pixel artwork, including clean gradients and particulars that would not be doable within the genuine medium.
Illustrations
Immediate: a costumed supervillain with fireplace powers trying down a road full of individuals whereas flying, overhead perspective, graphic novel illustration fashion of katsuhiro otomo, comedian e book fashion, jim lee, brian michael bendis
V1 by V3 understood perhaps 10% of my immediate (the “supervillain” half) however ignored every thing else. No background particulars, no correct perspective, and the themes themselves seemed like melted motion figures quite than menacing antagonists.
V4 and V5 confirmed vital enchancment however nonetheless missed essential parts that modified my immediate’s intent totally. V4 had the villain standing on the bottom as a substitute of flying, whereas V5 created a formless blob of fireplace above a crowd quite than a costumed character. The angles have been improper, the views confused, and the comedian e book fashion inconsistent.
Then got here V6 and V7, which nailed each single facet of the immediate. The overhead angle, the detailed road full of individuals, the flying supervillain with fireplace powers, and the distinctive comedian e book artwork fashion — they have been excellent. The pictures seemed like they have been torn straight from the pages of knowledgeable comedian e book, with dynamic poses, dramatic lighting, and that excellent steadiness of realism and stylization that defines the medium.
General Ideas?
After trying by all seven variations, one factor is evident: the largest leap in Midjourney’s evolution occurred between V3 and V4. That’s the place it lastly stopped being “simply fascinating tech” and began changing into an precise artistic instrument. Faces turned human, objects seemed like what they have been purported to, and prompts really meant one thing.
The leap from V6 to V7, although? It’s not as dramatic, however it’s nonetheless extremely spectacular. V7 tightens every thing. Lighting is extra pure. Scenes really feel extra lived-in. It’s like they smoothed out the previous couple of bumps that V6 couldn’t fairly polish. So, no, V7 isn’t a revolution — however it’s a refinement.
The place does it go from right here? Who is aware of. However one factor’s for certain: V1 walked so V7 may run with two legs.
