Up next
But how do AI images and videos actually work? | Guest video by Welch Labs 38 min 
Are you one of the lucky 50% that experience it? 2 min 
What If You Just Keep Digging? 11 min 
The Second Punic War - OverSimplified (Part 2) 38 min 
The Hidden Engineering of Niagara Falls 16 min 
Sawing a Dam in Half (on Purpose) 21 min 
Учусь юзать тело 🕺| Baby Steps 22 min 
The MAGNET Suspension SKATEBOARD 23 min 
The Windows Laptop Problem 11 min 
Заперт с Монстром 🤣 | Murky Divers 17 min 
How a broken heart affects your body 6 min 
Have I built the most amazing door Ever? 28 min 
I'm Actually Short 1 min 
The Bizarre Bases of Antenna Towers 19 min 
Amazing LEGO Tea Bag Machine 1 min 
Crash Course Geology Preview 3 min 
The illusion of the ‘self’ | The Gray Area 43 min 
The Hidden Engineering Behind the Falkirk Wheel 16 min 
Psychedelics, mind control, radiation-eating: The secret superpowers of fungi - Max G. Levy 7 min 
Men Transform Abandoned French House Under $80,000 | Start to Finish by @petitcopeau 
Name That Infrastructure (Ep. 14) #shorts 1 min 
Name That Infrastructure (Ep. 9) #shorts 1 min 
Kling 2.6 generates clean cinematic and highly consistent videos on higgsfield #aivideo 1 min 
Man Builds Amazing WORKSHOP With Garden In His Backyard | Start To Finish By @machikane_base 
The Hidden Engineering of Runways 19 min 
Recreating an Ancient Pump (with no moving parts) 14 min 
The Blob’s Epic Practical Horror Finale | The Blob 7 min 
Five things you need to know for the 2026 midterms | America, Actually 31 min 
Why you're sick of rainy days #science #SciShow #sick #biology #illness #rain 1 min 
I tested Dream Tech with my Biggest Fans! 15 min 
Olympic weightlifting, explained 9 min 
Lindsey Clancy and the surprisingly simple solution to helping moms with postpartum depression 2 min 
Why is everyone suddenly neurodivergent? 8 min 
The Best TINY Tech on the Internet! 20 min 
What The Heck Is Tuberculosis? #science #scishow #medicine #johngreen 1 min 
The Insane Engineering of Europa Clipper 21 min 
Have you HEARD this?? 2 min 
Why You Should Get Your DNA Sequenced After You Die 10 min 
Fixing the Most Dangerous Dam in the World 23 min 
not all sleep is equal 1 min 
This magnet can invert your brain, and it’s saving lives 3 min 
Crash Course Hosts Read Comments 2 min 
The Ring of Fire 2 min 
What is sumo, and why is it so popular? - Lee Thompson 6 min 
Hybrid Power Systems! | Real Life Power Armor (Part 4/6) 23 min 
The Hardest Problem AI Ever Solved, with Google DeepMind CEO 66 min 
Why Don’t We Use Mothballs Anymore? 12 min 
The Power Armor Legs Are Complete! ⚡| Real Life Power Armor (Part 7/6) 27 min 
What’s Ruining Our Ruins? 12 min 
Peppa's Sandcastle DISASTER ⛱️🌊 #PeppaPig #Shorts #Summer #LEGODUPLO 1 min 
We finally know why it's so good for you 7 min Diffusion models, CLIP, and the math of turning text into images Welch Labs Book: https://www.welchlabs.com/resources/imaginary-numbers-book Sections 0:00 - Intro 3:37 - CLIP 6:25 - Shared Embedding Space 8:16 - Diffusion Models & DDPM 11:44 - Learning Vector Fields 22:00 - DDIM 25:25 - Dall E 2 26:37 - Conditioning 30:02 - Guidance 33:39 - Negative Prompts 34:27 - Outro 35:32 - About guest videos Special Thanks to: Jonathan Ho - Jonathan is the Author of the DDPM paper and the Classifier Free Guidance Paper. https://arxiv.org/pdf/2006.11239 https://arxiv.org/pdf/2207.12598 Preetum Nakkiran - Preetum has an excellent introductory diffusion tutorial: https://arxiv.org/pdf/2406.08929 Chenyang Yuan - Many of the animations in this video were implemented using manim and Chenyang’s smalldiffusion library: https://github.com/yuanchenyang/smalldiffusion Cheyang also has a terrific tutorial and MIT course on diffusion models https://www.chenyang.co/diffusion.html https://www.practical-diffusion.org/ Other References All of Sander Dieleman’s diffusion blog posts are fantastic: https://sander.ai/ CLIP Paper: https://arxiv.org/pdf/2103.00020 DDIM Paper: https://arxiv.org/pdf/2010.02502 Score-Based Generative Modeling: https://arxiv.org/pdf/2011.13456 Wan2.1: https://github.com/Wan-Video/Wan2.1 Stable Diffusion: https://huggingface.co/stabilityai/stable-diffusion-2 Midjourney: https://www.midjourney.com/ Veo: https://deepmind.google/models/veo/ DallE 2 paper: https://cdn.openai.com/papers/dall-e-2.pdf Code for this video: https://github.com/stephencwelch/manim_videos/tree/master/_2025/sora Written by: Stephen Welch, with very helpful feedback from Grant Sanderson Produced by: Stephen Welch, Sam Baskin, and Pranav Gundu Technical Notes The noise videos in the opening have been passed through a VAE (actually, diffusion process happens in a compressed “latent” space), which acts very much like a video compressor - this is why the noise videos don’t look like pure salt and pepper. 6:15 CLIP: Although directly minimizing cosine similarity would push our vectors 180 degrees apart on a single batch, overall in practice, we need CLIP to maximize the uniformity of concepts over the hypersphere it's operating on. For this reason, we animated these vectors as orthogonal-ish. See: https://proceedings.mlr.press/v119/wang20k/wang20k.pdf Per Chenyang Yuan: at 10:15, the blurry image that results when removing random noise in DDPM is probably due to a mismatch in noise levels when calling the denoiser. When the denoiser is called on x_{t-1} during DDPM sampling, it is expected to have a certain noise level (let's call it sigma_{t-1}). If you generate x_{t-1} from x_t without adding noise, then the noise present in x_{t-1} is always smaller than sigma_{t-1}. This causes the denoiser to remove too much noise, thus pointing towards the mean of the dataset. The text conditioning input to stable diffusion is not the 512-dim text embedding vector, but the output of the layer before that, [with dimension 77x512](https://stackoverflow.com/a/79243065) For the vectors at 31:40 - Some implementations use f(x, t, cat) + alpha(f(x, t, cat) - f(x, t)), and some that do f(x, t) + alpha(f(x, t, cat) - f(x, t)), where an alpha value of 1 corresponds to no guidance. I chose the second format here to keep things simpler. At 30:30, the unconditional t=1 vector field looks a bit different from what it did at the 17:15 mark. This is the result of different models trained for different parts of the video, and likely a result of different random initializations. Premium Beat Music ID: EEDYZ3FP44YX8OWT
Packages published by the creator behind this video.
This creator has not set their rates yet. Send a request and we will pass it on — the price is agreed before anything is charged.
A recurring seat in my content. Subscribers enter a queue — or a random draw when there are more subscribers than slots — and appear in the video itself, in the credits, or in the description.
per month
Price on request
A guaranteed appearance of at least 10 seconds in my next video. Price depends on where in the runtime the insert lands.
Price on request
A guaranteed appearance of at least 10 seconds inside a Short cut from one of my existing videos.
one-off
Price on request