MIDJOURNEY & RUNWAY DROP HUGE UPDATES.

Fam,

I was traveling this weekend…so I’m late.

But I didn’t forget about you…and we got lots to talk about.

Big updates all around.

Here’s what’s going on today…

1. MIDJOURNEY: NEW EDITOR FEATURE?

2. RUNWAY: INDUSTRY CHANGING “ACT-1”

3. RUNWAY: TRANSITIONS WORKFLOW.

4. CUSTOM AI WORKFLOWS: NOW AVAILABLE.

5. MINIMAX + FLUX: ROBOTAXI LORA-TO-VIDEO.

6. PODCAST: BILLY BOMAN + COMMERCIAL PRODUCTION.

7. BEYOND MJ: NEWS ROUNDUP.

8. MJ OFFICE HOURS.

LET’S BOOGIE…

MIDJOURNEY: NEW IMAGE EDITOR?

Midjourney just made the BIGGEST CHANGE in recent months.

They just released their external image editor (it's available now...but...)

CHECK IT OUT ↓

HERE'S WHAT YOU NEED TO KNOW:

• The Midjourney Image Editor: lets you modify, expand, and reimagine imagery uploaded from your computer.

• Use Images from Outside Midjourney: Editing is done through text prompting and region selecting. It's compatible with model personalization, style references, character references, and image prompts.

• Retexture Images: You can use 'image retexturing mode' to keep the shape the same, but modify lighting, surface materials, and texture. (Very similar to ControlNet)

WHO CAN ACCESS IT:

• Members with yearly memberships.

• Members who have been subscribers for the past 12 months.

• Members with at least 10,000 images.

(I had to update this last minute…because MJ drops things at the most random times…more tests to come).

RUNWAY: INDUSTRY CHANGING “ACT-1”

Unless you swore off social media…you saw this drop.

Runway announced their NEW “ACT 1” Tool…which allows you to generate expressive characters based on source videos.

IF YOU MISSED IT → FULL POST

WHAT IT IS:

• You input a video of an actor (or yourself) talking

• RW generates complex facial animation

• NO traditional rigging or motion capture is necessary.

WHY IT’S IMPORTANT:

Dialog and natural character movement is lacking across all the platforms. If the test videos are as accurate as the generations will be…this will be really impressive.

WHEN TO EXPECT IT:

RW said it will begin rolling this out…meaning we’ll probably have it at sometime next week. So keep you’re eyes open.

MORE INFO → HERE

RUNWAY: TRANSITIONS WORKFLOW

Runway’s Keyframes opens up some creative oppos.

But it’s all contingent on the base images and how you direct the tool.

IF YOU MISSED IT → FULL POST

PROCESS:

1. Gen base images in MJ

2. Remix/Edit in MJ

3. Animate in Runway

MJ PROMPT EX: editorial photo, epic side profile shot of a child riding a bike through the streets of Rome, contrasting colors, spinning wheels, deep textures --ar 5:3 --p --stylize 200 --v 6.1

MJ REMIX PROMPT EX: editorial photo, epic side profile shot of a man riding a insane motorcycle through the streets of Rome, spinning wheels, contrasting colors, deep textures --ar 5:3 --p --stylize 200 --v 6.1

RUNWAY PROMPT EX: child riding a bicycle to the right as he seamlessly transforms into a grown man riding a motorcycle, 35mm style

QUICK THOUGHTS:

• To enhance motion use EDIT in MJ

• To change the perspective/composition

• CHILD IMG → Positioning the BIKE in bottom left

• MOTORCYCLE IMG → Positioning the car in the bottom right

• It gives runway some room to fill in the blank

IF YOU WANT MORE OF THESE → CHECK THIS OUT

CUSTOM AI WORKFLOWS:

We’ve been cooking in the lab at SUPERSIDE.

Seems like each week we’re working on something new…and most of the time, these custom workflows end up becoming productized.

Like this “Breif-to-Email Marketing Campaign” workflow for Vimeo ↓

WHY?:

• 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.

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

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

MINIMAX + FLUX: ROBOTAXI LORA-TO-VIDEO

What Midjourney lacks is the ability to train LORA models. (For now)

So I took CIVITAI for a ride to test the FLUX LORA’s everyone’s talking about.

CHECK IT OUT → FULL POST

WHY IT’S IMPORTANT?

• If you want product/object/character consistency…FLUX has made that easy.

• This whole process took less than 30 mins.

• And MINIMAX is excelling at fast action motion.

You can also upload different LORAs and try ones from the community.

For a more comprehensive training guide check this out ↓

PODCAST: BILLY BOMAN + COMMERCIAL AI PRODUCTION

Drew and I are back and we welcomed the great and powerful Billy Boman

If you’re into AI video and how this being used commercially…

CHECK THIS POD OUT ↓

HIGHLIGHTS:

• [00:00] Creative Journey [03:45] Career Shift [08:00] Midjourney Impact [19:00] AI Teamwork [24:30] Fluid Storyboarding [31:00] Achieving Emotion w/ AI [39:15] Innovative Thinking [44:45] AI Assets [51:30] Legal Landscape [57:45] Creative Amplifier [01:02:00] Rapid Advancements [01:04:45] Production Revolution

If you like the episode…drop us a “LIKE” or “SUBSCRIBE” (it really helps)

Respond to this email if you have any feedback (it’s all helpful).

MIDJOURNEY OFFICE HOURS (10/16):

Video Models

• Training progressing well

• Challenges: Speed, quality, and cost

• Goals: Improve speed and quality, manage expenses

• Transitioning from research to productization

3D Models

• Delayed due to team health issues

• Expected to be ready soon

V7 Model

• Training ongoing with multiple restarts

• Transitioning to tuning mode soon

• Enhanced text understanding, speed, and coherence

• Trained on more data, introducing new qualities

Depth Reference (Retexture)

• Alters textures while maintaining shapes

• Acts as photo filter in editor

• Exclusive to editor, not create flow

Storytelling Tools

• First storytelling features introduced

• Videos and talks by team members

• Aim to make tools accessible to all users over time

Community Features and Feedback

• New system for submitting and rating questions

• Top-voted questions addressed in office hours

• Developing better Explore experience

• Personalized Explore aiming to be 10x better

Deepfake Prevention

• Banning users who attempt deepfakes

• Focus on discouraging deepfake creation

Have fun.

If you need help…DM me on Linkedin.

Until next time…

Rory