⚡ Texturized 100% with AI in 5 minutes 🤯

I remember spending 30 hours on texturing, lighting, and rendering to finish a render like this. And now, anyone can do it in seconds. Everything is going to change!

Here's a step-by-step tutorial 🧵👇

1/ First, the transformation, using the brand new Magnific Style Transfer:

I used a reference image with the style I wanted to transfer to the base render and increased Structure Strength up to 100%.

IT'S FINALLY HERE!

🔥 Magnific Style Transfer! 🔥

Transform any image, controlling the amount of style transferred and the structural integrity ✨ Infinite use cases! 3d, video games, interior design, for fun...

Details, 10 tutorials & HUGE surprise 🧵👇 If you can't wait, here's the link to that website that most of you already love (and we love you back! ❤️).

LINK 👇👇👇

magnific.ai

⚡ Isn't this something?

It seems that @juliewdesign_ has perfectly cracked the parameters for perfect Portraits with Magnific!

Here's a TUTORIAL that emerged from a chat with her. She has allowed me to share her secret mojo with you 😉👇 Midjourney prompt: "young blonde woman in blue shirt with big, in the style of soft and rounded forms, avacadopunk, webcam photography, , dark teal and light gray, dutch marine scenes, detailed facial features --ar 57:71 --v 6.0"

Magnific parameters:

⚡ This was sci-fi until today!

We cracked super reimagined hi-res upscaling!

Only for beta testers for now, in 24 hours for everyone!

- Output size up to 10,000 x 10,000 pixels. Any Midjourney image to 8x! 🤯
- New slider: "Fractality"
- Zoom in/out results using "Z + wheel" Tomorrow, if the GPUs hold up today 😅, I will prepare a more detailed thread with lots of tutorials and examples for the official launch!

🔥 The Force is Strong With Him 🔥

100% AI Generated. Step by step tutorial:

1⃣ Midjourney: "Film still, hipyer realistic, intricate, detailed, Donald Duck as a jedi with a light saber, dark, glowing lightsaber --v 6.0 --ar 2:3 --style raw"

2⃣ 1st Upscale in Magnific:
- Preset: "Films & Photography"
- Prompt: "Film still, hipyer realistic, intricate, detailed, Donald Duck as a jedi with a light saber, dark, glowing lightsaber"
- Creativity: 2
- HDR: 1
- Resemblance: -1

3⃣ 2nd Upscale in Magnific:
Reduce Creativity: 1 and HDR: 0.

Share this post, and I'll be inspired to post more tutorials! Thank you! 💌🙏 The most amazing thing about this is that Disney could produce this movie: they have al the IPs!

⚡ Many can enter a prompt and see what happens, but few truly master getting the most detail from AI creations.

I learned how to create 4K resolution anime-style bucolic scenes inspired by Studio Ghibli 🤯

💡 Using Midjourney + Magnific AI

Step by step tutorial 👇

1⃣ First step, is generating the raw images in Midjourney.

First image: "a green house by studio ghibli that has flowers on top, in the style of mark keathley, ueda fumito, nostalgic paintings, swiss style, colorful sidewalk scenes, high-contrast shading, jacek malczewski, studio ghibli --ar 51:64 --niji 5 --style expressive"

Second image: "a green house that has flowers on top, in the style of mark keathley, ueda fumito, nostalgic paintings, swiss style, colorful sidewalk scenes, high-contrast shading, jacek malczewski --ar 51:64"

Don't worry about the low detail, we will fix that in following steps.

⚡ I will include in this thread all the Magnific tutorials.

We will upload all of this to a comprehensive knowledge base on the website eventually, but in the meantime, I want to have a quick link.

I will also include tutorials from YOU. So if you have one, let me know!

https://twitter.com/javilopen/status/1729845570792304644

⚡ TUTORIAL: Creating photorealistic portraits with AI

Lots of people are generating AI images, but only a few really know how to achieve this level of realism.

Here are the prompts and a step-by-step tutorial (plus a surprise at the end of the thread):



1. First, create the base images in Midjourney.

For this, you'll need a really solid prompt, especially for portraits!

Here are you have the three AI generated images in Midjourney along with their prompts in the ALT text, so you can learn from them!

⚡ OK, this is impressive 🤯

I also wanted to try upscaling Lara Croft from PS1 to super high resolution! Mini step by step tutorial 👇

1/ First Upscaling:

- Preset: Standard
- Creativity: 3
- HDR: 0
- Resemblance: -1
- Prompt: "Screenshot, (detailed eyes:1.2), (detailed lips:1.1), (young cute:1.2), Lara Croft, beautiful, 8k"

2/ Second Upscaling:

Use the same settings, but change Creativity to 1.
Original image VS Reimagined upscaling. Holly molly!

FIRST MAGNIFIC AI 🪄 TUTORIAL

Midjourney + Magnific = 🔥 INSANE HIGH RES 🔥

← Left: original Midjourney image.
→ Right: upscaled with Magnific (4x).

Step by step tutorial 🧵👇

A lot of people have been asking for this tutorial because it's hard to believe this is real 😂, but it is!

Link to Magnific Upscaler & Enhancer:

Check out the small details added! magnific.ai

WE ARE SO BACK!!!

🔥 Finally! We are launching Magnific AI 🔥

The image Upscaler & Enhancer that feels like Magic 🪄

Link & details 🧵👇 If you can't wait, here's the link. Once you're on the website, click on "Upscale an image" to register.

Ah! And you can follow @Magnific_AI for the future tutorials I'll be posting there.

LINK 👇👇👇

magnific.ai

⚡ A question I'm often asked is:

WHICH AI IMAGE GENERATOR IS THE BEST?

Midjourney
DALL·E 3
Stable Diffusion
Leonardo AI
Adobe Firefly
Ideogram
Or another?

Honestly, it really DEPENDS. Each one might be the best in its own specific VERTICAL.

Here's my take on it 🧵👇 MIDJOURNEY

In my opinion, Midjourney is the most VERSATILE. It performs well in almost any situation (except for writing texts), and its images are usually very beautiful. As of today, it remains my favorite.

So, Generative AI in REAL TIME is here 🔥

Only 4 days have passed and nothing will ever be the same again. All the design tools, all the work processes, EVERYTHING is going to change.

💡 Here is everything you need to know about LCM-LoRA 🧵👇

So, 4 days ago. YES, four days ago. This paper appeared and KABOOM!

🔗 LINK: arxiv.org/abs/2311.05556

This is the video AI Haters tried to cancel, created by Martin Nebelong, who has just given me permission to republish it.

He felt compelled to remove it due to the incredible amount of bullying he was receiving. In his words: "The amount of hate, negativity, and fear was staggering, and I didn't want to defend myself against that."

Nothing outrages me more me more than the "cancel culture" and the unfair pressures on creators who are simply using AI tools and showing the world how they can learn these new creative processes.

His video is amazing. It shows how Generative AI nowadays is much more than "just writing a prompt". And how it will enable artists and designers to experiment in countless new ways.

Remember that it's not that artists/designers hate AI! AI Haters are just a small group of people and they don't speak for all artists. In fact, I know many more artists and designers who already use it daily. But they just don't make noise: they focus on learning and creating.

Martin, you, and any other creator using AI, has all my support!
For this video, @MartinNebelong used Photoshop captured and turned to a Webcam that then feeds into that AI model. He is working in making short guide!

And here you have a different test: Dreams on PS5 with a layer of AI on top!

https://twitter.com/MartinNebelong/status/1722383172871049272

Midjourney is great, but the amount of detail you can achieve with ✨ Leonardo AI ✨ it's insane 🤯

Mini tutorials with FULL pipeline + details 🧵👇 🔗 Leonardo AI (tool):
💬 Prompt: "Close up portrait, cute, baby dragon photo shoot, medieval".
🤖 Parameters: Alchemy v2 + Res: 1360x768 + Leonardo Vision XL + Photography preset

If you have any question, feel free to ask! leonardo.ai

Midjourney, DALL•E 3 and GPT-4 have opened a world of endless possibilities.

I just coded "Angry Pumpkins 🎃" (any resemblance is purely coincidental 😂) using GPT-4 for all the coding and Midjourney / DALLE for the graphics.

Here are the prompts and the process I followed: First of all, would you like to play the game?

Here's a link! (Currently, it doesn't work on mobile):

If you read the text below the game screen, which provides explanations, you'll see how you can create your own levels and play them! :) bestaiprompts.art/angry-pumpkins…

⚡️ Well, isn't that something?!

Did you know we can specify not only the ASPECT RATIO but also the SEED in DALL·E 3 generations?

And we can use that for CHARACTER CONSISTENCY!

And more! I made ChatGPT show me exactly how it communicates with DALL·E 3. Check it out 🧵👇 The first thing I wanted to know was exactly which parameters, besides the prompt, we can REALLY send to DALL·E 3.

Sadly, we can forget about the "stylize" options and such that come with, for instance, Midjourney. ChatGPT would just send them as part of the prompt.

But we can:

⚡ Decoding X's (Twitter) ALGORITHM

X released part of its source code months ago, but very few have truly dived into its depths.

I've read everything that has been published and studied the source code so you don't have to.

In this thread, I'll unveil ALL its secrets 🧵👇 - How to go viral on X?
- Why do some posts go viral while others don't?
- Does each account have a scoring/karma rating?
- What raises or lowers that karma?
- How many of my daily posts can be successful?
- Pickles on your burger or no pickles?

⚡️ Well, isn't that something?!

In a paper published by Google, they explain that there is a phrase that systematically increases the accuracy in the answers (+71%) of their LLM (PaLM 2):

💬 "Take a deep breath and work on this problem step by step"

Works on ChatGPT too!

🔗 LINK (paper):

Remember: "Take a deep breath and work on this problem step-by-step."

😉😉😉arxiv.org/pdf/2309.03409…

🔴 OK, this is so Black Mirror in real life that it gives me the creeps 🤯

Let me introduce you to Billie (aka Kendall AI). An avatar fully powered by AI!

Meta AI is taking very recognizable celebrities and turning them into characters with different names on Instagram & Facebook.

Meta AI has reportedly paid outrageous sums of money to celebrities to turn them into characters with different names. Specifically, according to The Information, Meta would have paid up to five million dollars to one of these stars for just six hours of work.

The result? Mass confusion: users don't understand why they see Kendall on their timeline claiming her name is Billie 🤣

Ladies and gentlemen, fasten your seatbelts because we're headed to the Uncanny Valley!
The AI-controlled avatar posing for you all. No, this isn't Kendall, it's her eternal digital version.

BLACK MIRROR.

🔥 15 AI updates that have blown up in the last few weeks 🔥

I'm back and trying out all the AI updates that I've missed during these 3 weeks of travel. THERE ARE SO MANY!

🦄 Starting with DALL·E 3! Prompt: Batman in a small car saying: "I'm back!"

🧵 Let's see them together!
1. DALL·E 3

Tremble Midjourney, tremble!

A big surprise for many. Just when we thought OpenAI had thrown in the towel on image generation topics, they pull a new version of Dalle out of the hat!

And honestly, it rocks:

- It is capable of writing legible texts (or at least, like Ideogram, most of the time).
- Great coherence and understanding of the prompt.
- Higher image quality: for illustrations, it is truly spectacular.

💡 Where to try it?

- On Bing Create (where I am trying it):
- In Bing Chat (when it wants to, it fails these days because it is overloaded).
- In Chat GPT Plus (they are gradually giving access).

Not bad at all!