Varient3 retweeted
I think @elonmusk owes Apple an apology. I won’t say this is perfect (hackers will inevitably find problems), but it seems to me to be quite thoughtfully conceived, and likely is state of the art.
Varient3 retweeted
😯 AMAZING: Did you know the Aerostat was designed to endure the harshest conditions, including this sudden Dust Devil!
Nothing can stop #WorldMobile!
Varient3 retweeted
Yesterday, this big #Cardano transaction moves 1.3B $ADA (around $837,000,000) and it cost them $0.5 in fees
eutxo.org/transaction/bcaabc…
cexplorer.io/tx/bcaabc2a9c8f…
Varient3 retweeted
✨ For the last few months I have been reverse engineering Magnific AI's famous upscaler
It uses MultiDiffusion, ControlNet tiles and details LoRas
In true AI spirit, I am open sourcing it for everyone to use for free in your apps
It's designed to easily control the amount of hallucination, resemblance and creativity.
You can use it on Replicate now: replicate.com/philipp1337x/c…
Or run it locally with A1111'a web UI (parameters are in the comments)
The code is fully open source: github.com/philz1337x/clarit…
Varient3 retweeted
Found 12 signed 1st edition copies of INTO THE INFERNO in my studio! Bundling them with an option for an 8x10 print (selections below) for $125. You basically get the book for free. When they're gone, they're gone! Link below too.
Varient3 retweeted
Introducing Custom Characters at bedtimestory.ai
Put your kids at the center of a story.
Custom characters allows for consistent looks, backstory and more.
🧵 A thread
Varient3 retweeted
After NeRFs and Gaussian Splats, it looks like we're back to meshes!
Binary Opacity Grids are mesh-based views that are able to capture fine geometric detail which can be rendered in real-time on mobile devices.
binary-opacity-grid.github.i…
Varient3 retweeted
👀
LetzAI V2 will allow brands to train realistic AI models of their products and make those models available to the public ✌
A new era for user-generated content.
Here are some unedited examples I made just now. Prompt links for V2 in ALT.
ALT https://letz.ai/image/1042b2b3-7266-4814-b1e5-ec48a3413e16
ALT https://letz.ai/image/c15719e3-a384-4562-95b2-7a8e8ccd28fd
ALT https://letz.ai/image/3148c3f7-b3be-4c3a-9c63-ed04a125cd45
ALT https://letz.ai/image/82cc187a-2dfe-4e63-aed2-faa5f3db6229
Varient3 retweeted
Good Morning!
What I saw vs What I felt:
📍 Sunset on New Years Day at the Devil’s Punchbowl in Oregon
❤️🔥 The end of a beginning. In 2019 I learned that traveling doesn’t take you away from your pain. It travels with you. When I photograph “comfortable isolation” it is not to “get away” it is to silence the discord.
I’m searching for moments within scenes that feel like portals into yourself. My escapes are not “away from” a place or situation but “into the self”, beneath the noise that crowds the surface of our thoughts and attention.
It just so happened that a litterally portal opened in Oregon for me this evening.
collected by @GuyNorcal
Follow for my staycation photography 😂
Varient3 retweeted
"AI generated images can't be art", "Learn how to draw", "AI art is the lowest of lowest art-forms".
Meanwhile, I'm sculpting on my PS5 in Dreams combined with a realtime AI transformation layer, feeling that I have caught a sliver of what building immersive worlds will be like in the very near future.
The tone in the scene is darker than usual.. the state of the world and the way we treat each other gets to me now and then despite a (maybe naive) belief in the beauty and good we're all capable of.
#thepowerofdreams #ai
Varient3 retweeted
So it turns out that OpenAI's Sora is even more insane than I thought. Here are some of the absolutely nutty things it can do ⬇️
Varient3 retweeted
Announcing Sora — our model which creates minute-long videos from a text prompt: openai.com/sora
Varient3 retweeted
How is this generated?
These videos look more and more like they are exact clips from the training data.
I know this is not the case
But it’s so realistic that it’s spooky
h/t @rrhoover
Varient3 retweeted
It’s actually fuckin crazy to me that Midjourney generated this image from a single word
The prompt was “omniscient —s 1000”
Varient3 retweeted
🚨 Google just dropped Gemini Pro 1.5, the next version of its AI model with a 1,000,000+ token context length.
The model can now understand entire books, full movies, and podcast series' all in one go.
This surpasses all other competitor Chatbot context windows by a longshot.
Gemini Pro 1.5 now uses a Mixture-of-Experts (MoE) approach for more efficient training and higher-quality responses.
Impressive to see Google ship Pro 1.5 mid-February just after the launch of Pro 1.0 in December, and Advanced last week.
Varient3 retweeted
As you know, my explorations of the Gen AI space is ultimately all about creative control. You should be able to shape the generative matter using all your artistic sensibilities and your aesthetic sense.
OpenAI's Sora is a huge technological leap, but what excites me the most about it is the modalities where it depends on input other than text alone. Such as video to video. Here's an example of how Sora can change an input video.
Base video🧵