MIDJOURNEY V7 RELEASED + RUNWAY GEN-4 RELEASED (1)

Fam,

The time has come…Midjourney V7 is HERE…

And it’s good (idc what anyone says).

As if that wasn’t enough…

Runway released GEN-4 also….so my brain hurts.

Here’s what we got today:

1. MIDJOURNEY: RELEASES VERSION 7

2. RUNWAY: RELEASES GEN-4

3. CHATGPT: CREATING FONTS

4. CHATGPT: SOOO MANY USES CASES

5. SUPERSIDE: DONE-FOR-YOU CREATIVE

6. PODCAST: DOUBLE HEADER

7. MJ OFFICE HOURS

LET’S BOOGIE…

MIDJOURNEY: RELEASES VERSION 7

As is tradition…Midjourney released V7 under the cover of darkness. They were under pressure given the current climate. (And they needed to get something out)

Now that it’s here…It’s powerful…it’s advanced…it’s raw.

HERE’S THE REVIEW → FULL POST

WHAT’S NEW (Per MJ):

1. Increased visual firepower

2. Improved prompt coherence

3. Improved textures + colors + depth

4. New “draft” mode for quick iteration

5. Voice-to-image capabilities. (Kinda cool)

6. READ FULL RELEASE HERE

INITIAL THOUGHTS:

• It’s definitely got some power

• But it has some rough edges

• They’re planning rapid updates

• So expect changes in the next few weeks

EX PROMPT: low-angle street style photo of a samurai in black attire, with neon yellow contrasting colors. dark environment, the camera flash illuminates the samurai making him look over-exposed --chaos 5 --ar 5:6 --profile 7pm5s7i --stylize 500

TIPS FOR V7:

• Personalization (--p) is the key variable w V7

• Spend time training V7 personalization

• It makes the model 10x better

• I ranked about 1k images in V7

• You can use your old --p codes

• But they’re not tuned to V7

• The minute I started training…Instant improvement

• There’s better coherence on "Relax Mode" vs Turbo.

EX PROMPT: cinematic still, author writing at a desk as massive cybernetic smooth creature skull hangs above his workspace, brilliant intricate details --chaos 5 --ar 3:4 --stylize 750 {--p,}

OVERALL THOUGHTS:

• This is a new architecture (allegedly)

• Meaning…prompting will be different

• Need some time to work through this 

• Textures + depth very improved

• I don’t think any model can match the depth

• Really like the new aesthetic model 

• MBs and SREF codes might need a minute

First imp…visually excellent + not totally polished.

SO many more things to test.

MIDJOURNEY MASTERY: BIG COURSE UPDATE

Made some BIG updates to Midjourney Mastery.

Adding over 1.5+ HOURS of NEW VIDEO CONTENT on...

• Personalization (--p): Usage and Strategy

• Midjourney Moodboards: Usage + Strategy + Workflows

• Exclusively in the Midjourney Mastery Course.

As a little thank you for all the support here's the deal.

>>$47 OFF MIDJOURNEY MASTERY<<

Dropping the price from $187 to ($140) for the next 72 hours.

If you haven’t joined the party ↓

RUNWAY: RELEASES GEN-4

This one was on the horizon…and it came earlier than expected.

Runway still has some of the most useful features in Ai Vid.

And now they’re going to be rolling out a lot more.

IF YOU MISSED IT → CHECK IT OUT

WHATS NEW: 

• Currently only I2V. 

• Improved Aesthetics

• Improved prompt coherence

• Dynamic subj + camera motion

• Multiple aspect ratios 

There’s a lot to break down here.

But the most impressive thing is the subject + camera motion improvement.

RW PROMPT STRUCTURE: [shot type], [subject + action + motion + direction] [env]. [Camera motion], [env description], [lighting], [mood/emotion].

RW PROMPT EX: A fast action tracking shot of a young man skateboarding towards us down a dusty desert road. The camera pulls out on him as he weaves on his skateboard casually. Behind him, a colossal fiery mushroom cloud expands into the sky and towards us, growing methodically larger. The lighting is natural and cinematic, with warm tones from the blast contrasting against the cool shadows of the dunes. The handheld camera movement adds 

OVERALL THOUGHTS:

• Very nice update

• Frames images work very well w G4

• RW is building a full suite of tools

• Gens typically take 2-3 mins 

• Sometimes the output is soft

• But that can we corrected w input images.

Wayyyy more testing to come on this as they roll out new feature.

CHATGPT 4o: CREATING FONTS

ChatGPT’s image model has been hard to ignore.

It’s like a creative co-pilot.

My advice…just start asking it to do tough things…(it can)

FOR EXAMPLE → IT CAN CREATE FONTS

PROCESS:

1. Provide ChatGPT reference image.

2. Ask for it to create a font.

3. Upload image.

4. Ask to test font on image.

So I figured i’d do this with my logo…here’s how it went…

(MORE DETAILS IN THREAD)

STEP 1: CREATE REF IMAGE

STEP 2: ASK CHATGPT TO CREATE A FONT

PROMPT: this is the letter S from my logo. Do we think we can build an entire font from this?

STEP 2.5: ITERATE FONT FOR DIFFERENT WEIGHTS

STEP 3/4: UPLOAD AN IMAGE + ASK FOR MOCKUP

FONT BUILDING:

This works great within ChatGPT..but if you want to export the font to use it in PS or Figma you’ll need to do some extra work.

I’ll do another piece of content on that. (Stay tuned)

CHATGPT 4o: SO MANY USE CASES:

If you can’t tell…I can’t get enough.

So I’m dropping a few more USE CASES we’ve been playing with.

HOW ABOUT → GIF ADS OR EMAIL/SMS ASSETS

(Full Details in Thread Below)

CUSTOM LUTS:

This was a fun one for anyone playing in the ai video space.

You can give ChatGPT a reference image to create a LUT.

So you’re video projects are color graded specifically.

CHECK IT OUT → FULL POST

CUSTOM AI WORKFLOWS: DONE FOR YOU

Never a dull moment at SUPERSIDE.

We’ve now surpassed 2000+ PROJECTS completed with Ai

RESULTS: 50-85% Reduction in Hours → $1.4 Million Saved for Clients.

For brads like Amazon, Reddit, Google, Coinbase, Booking.com, etc…

WHY USE AI?:

• So we get from brief-to-first-draft as rapidly as possible.

• With key visuals/copy/ux/ui trained and tailored to brand guidelines/standards

• Fully automated with tools like OpenAi/Claude/Flux/Figma

• So we cut the back-and-forth time on “does this fit the brand?”

• And we focus on strategy, visual firepower, and iteration.

The best part?…If we can’t do it with Ai…we have 500+ traditionally trained world-class designers…who can do it by hand.

Have a design problem that needs solving? (We like those)

It’s worth a convo (at the very minimum).

PODCAST: DOUBLE HEADER

It was a big week in AI…so Drew and I dropped (2) PODS.

Typical nonsense…with maybe a touch of brilliance?

01. MIDJOURNEY V7 RELEASE: OUR THOUGHTS

02. CHATGPT 4o IMAGE GEN: IMMEDIATE REACTION

HERE’S WHERE YOU CAN FIND IT:

YOUTUBE | SPOTIFY | APPLE

MIDJOURNEY: OFFICE HOURS

Version 7 (V7) Launch

• LAUNCHED

• Improved quality, better textures,

• Better coherence, and better prompt understanding.

• New interface.

Speed Modes

• Two speed modes will be available at launch: standard and draft.

• The standard mode focuses on quality.

• The draft mode is designed to be fast and more conversational

• Ideal for exploring different initial ideas.

Features Available at V7 Launch

• Style reference.

• Personalization.

• Variations.

• Negative prompts (--no).

• V7 will work with the V6 upscaler and the V6 editor.

Delayed Features

• CREF and Omni-reference will be delayed

• They require further adjustments.

• An improved style reference is in development,

• Along with a new random S-Ref and a style explorer.

• A new upscaler will arrive later.

• Remix and multi-prompts will not be available at launch.

Future Developments (Next Weeks)

• Weekly releases will occur over the next four weeks

• With at least one new feature added each week.