How it works, in a nutshell:
- User provides a topic
- Claude breaks it into sub-topics
- An agent with access to Google builds a report for each sub-topic
- A final Claude instance takes in all of the sub-topic reports, and generates a final, comprehensive report

Don’t worry everyone. There’s only one backdoor and we found it, so everything’s totally ok now.

Ride 5 in vr looks so amazing, I can’t stop playing #PCVR #VirtualReality #UEVR

I observe how gradually larger and larger percentage of my time in my software engineering role is spent on writing prompts for AI that writes code for me, rather than on writing that code manually myself. And it makes me comfortable working with languages I've never used before.

Wow. This is pretty impressive!! Achieved in mere seconds ⏱️

It's truly inspiring. Kudos to @tripoa1, Stability AI & Emad acc/acc for the outstanding work.

The OpenAI blue alien Chibi version is now in #3D 😎❤️‍🔥

Just in 👀

this is the most amazing audio2video I have ever seen.
It is called EMO: Emote Portrait Alive

Vasek Mlejnsky GitHub E2B They have openly admitted they are using automations and AI to suspend accounts, it’s insane. When I challenged them on this, the developer blocked me on X

The best way to learn is by doing.

GPT in 60 Lines of NumPy 😎

In this post, you will learn how to implement a GPT from scratch in just 60 lines of Numpy. Then you can load the trained GPT-2 model weights released by OpenAI into this implementation and generate some text. So

Prasann Pandya The final form for ChatGPT is not chat : )

The first public dynamic Gaussian Splatting implementation in VR is here from San Francisco based Gmix, compatible with WebXR and Apple Vision Pro.

🔗radiancefields.com/gmix-ai-dynami…

Fooling around with the QR CN.

Introducing Sora, our text-to-video model.

Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions.

openai.com/sora

Prompt: “Beautiful, snowy

Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.

LCM AnimateDiff has the best vid2vid realism that I have seen! This is a huge leap forward! #AnimateDiff #stablediffusion #aianimation

🚨Breakthrough

Generate entire worlds with Outtakes for  Vision Pro

Immersive animated 360 worlds upscaled to 8K, runs entirely local on Apple Silicon. All you need is a prompt and Outtakes AI will handle the rest.

Coming soon to the App Store. DM for beta.

The future of movie making

#genuary Day 15: Physics Library

Finally got around to trying out Mickey van Olst ’s ChainGPU library. Had so much fun tweaking the physics params on the fly and combining it with the TD MediaPipe plugin by Dom Scott and myself.

Created in #TouchDesigner
#genuary 2024

I got BCI tracking now! I was able to directly control the ears on my avatar with my MIND after some training in the mirror! Super cool stuff!

Thanks to Lethos Storms for the hardware and Chillout_Charles for the software!

Decided to make an even larger fantasy group shot using Magnific.ai’s 8x upscaling/refining. Final res: 11,648 x 6528. Closeups and original image below. Shout out again to Javi Lopez ⛩️ & Emilio for this epic product that opens up a lot of workflow options! 🧵👇1/7:

Can generative AI imagine what Alice saw in her journey in the Wonderland 🏞️🚶‍♀️? Introducing WonderJourney: Create a journey (a long sequence of diverse yet connected 3D scenes) from a single image or text! 🧵1/N

Web: kovenyu.com/wonderjourney/
arxiv: arxiv.org/abs/2312.03884