Creativity AI #4: A fresh perspective on the synergy of AI, creativity, and productivity
Midjourney V7 in mid-Feb, style code mixing, ChatGPT Tasks, Runway Frames, and SOUNDRAW V2
Creativity AI is a new publication that brings together interesting articles and recent developments in the AI world related to creativity and productivity.
A busy week
There have been many updates in the last few days. Almost all AI companies have something new to offer. It's challenging to keep up with the constant changes.
But, hey, stay calm and pick things you can work on and try out. One thing at a time. And if they do, we will learn about a newer feature. Have fun experimenting and seeing what you can come up with.
Midjourney Office Hours (2025-01-22)
Midjourney V7 may be released in mid-February. It may support "omni reference" for objects and logos references. It will include better character references.
The 3D feature may allow users to move the camera around the scene for reframing.
The video feature could be released on top of partner models, adding "Midjourney magic" to the creation. Midjourney wants to test the market first before allocating more resources.
Midjourney style codes mixing
Midjourney users can now combine Moodboard, Style Reference, and Personalization codes to create a new aesthetic.
Read this article to find out more about this new feature.
ChatGPT
You can now schedule tasks and receive reminders in ChatGPT.
This new beta feature is not enabled automatically. So you need to look for it in the Model menu at the top left and choose "GPT-4o with scheduled tasks."
Here's a simple example: I ask the bot to remind me to drink more water.
And after 5 minutes, I received an email requesting that I drink water RIGHT NOW! Quite aggressive. Okay, I am going to drink now...
Clicking on the Profile in the top right corner will also take you to the "Tasks" section.
So there's the task: remind me to drink water.
Of course, that isn't the only thing it can do. Although it is still in the early beta stage, it can already handle more complex tasks, such as creating recurring tasks, using custom instructions, and so on.
Why does this matter?
The Task feature can be useful for reminding you to do something at a specific time.
It could also be used to automate tasks such as content planning, daily skill development exercises, and daily summaries.
Runway
Runway introduces Frames, an image generation feature for Unlimited and Enterprise plan users.
Unlike Midjourney, Runway suggests using a long and detailed prompting style for better stylistic control. It also includes pre-made styles and keywords to help users get started creating images.
Frames generate four images per grid. It includes a Vary option for generating variations of a source image, but it cannot yet maintain character consistency.
I would have expected that using images generated by Runway's Frames feature would reduce the failure rate when creating videos. However, it still has a high failure rate in my testing, with no noticeable difference when using Midjourney images for video creation.
This is an example of an image created with Frames.
This video was created using the same image. (Many of them failed to stay coherent, but this one is okay.) But the cat is in mid-air doing strange things.
Why does this matter?
It is now easier to generate images without having to import them from Midjourney, especially if you want to create videos quickly.
The Frames may include additional features in the future, but for the time being, you have more control and options when using Midjourney to generate photographic images.
Perplexity
Perplexity introduces the Sonar and Sonar Pro APIs for developers to enable search capabilities.
Why does this matter?
Nothing for average users unless you are building AI apps.
Note that Sonar Huge (the other Sonar model) has similarities with Midjourney where it could be used to learn new prompt keywords for image generation in addition to the /describe command in Midjourney. Read this related article.
SOUNDRAW
SOUNDRAW just released its V2, which can combine two different musical genres to produce original, high-quality music that is royalty-free. It can be used as background music for videos, podcasts, social media etc.
The V2 is a significant improvement over the first version, with more varied and less repetitive music produced. Additionally, users can alter the choruses, intros, and other parts to suit the requirements of the project or video.
An example of a short background music clip that combines Techno & Trance and Orchestra genres:
Why does this matter?
Up until January 28, 2025, SOUNDRAW is currently offering a 50% discount ($8.49 per month for infinite download). If you have been waiting for sales to sign up for a platform that produces AI music, take advantage of this opportunity.
Similar to combining Midjourney codes, combining various musical genres results in something fresh and distinctively you!
Keyword discovery
"Solar photogram" is a keyword that will set your Midjourney image on fire, literally.
solar photogram of a warrior ready to fight --ar 16:9 --v 6.1 
And when I animate the images in Runway:
Inspiration
The Public Domain Image Archive is a website with over 10,000 out-of-copyright historical images available for everyone to explore and reuse. It's a fantastic resource for Midjourney users looking to experiment and create new styles.
This is the image I made, inspired by one from the website.
A fluffy ginger cat sleeping in a wicker basket lined with a plaid blanket. --sref https://s.mj.run/bXb_SG2HREg --ar 16:9 --v 6.1What is its reference image, you might ask?
The artwork is from the Japanese design magazine Shin-Bijutsukai (1902). You can find it here.
Have a look at the other images in the collection.
Create something fun and inspiring!
Recent articles
Free Geeky Animals' Sref code: 3053564100
Cover prompt: A close-up of a man staring into a robot's mirrored eyes, his reflection showing a contemplative expression. --ar 16:9 --sref 3053564100 --v 6.1
Invitation to publish on Creativity AI
Please contact me if you want to publish on Creativity AI.
This publication will be distributed through Substack and Medium platforms, reaching thousands of readers.
Email: animalsgeeky@gmail.com