Inayofuata
But how do AI images and videos actually work? | Guest video by Welch Labs 38 dak 
What's Under Your Feet in New York City? 23 dak 
Берегись Паука! 🕷️ (Fever Meme) 18 dak 
FORECAST 4 dak 
Abandoned in Germany with $0 for 24 Hours (Man vs. Woman) 46 dak 
Abandoned Disposable Camera Photos Revealed.. 1 dak 
Actually Stranded on a Deserted Island for 42 Hours (real emergency) 37 dak 
How I Made My 2025 Secret Santa's year THE BEST 16 dak 
72 Hours Lost in the World’s Richest Dictatorship (no hotel, no food, no transportation...) 38 dak 
Inside Pakistan’s Most Dangerous City (surreal) 38 dak 
The Giant Rotting Pink Bunny in the Middle of Nowhere | Fascinating Horror Shorts 1 dak 
Why Gravity Is A Lie, explained in Zero G 23 dak 
IBM’s mainframe chip collab, NVIDIA’s Poolside deal & Ox Alpha’s reveal 31 dak 
Overboard From A Cruise Ship | Fascinating Horror 12 dak 
How the US' Only District Works 19 dak 
This Poodle Killed Three People | Fascinating Horror Shorts 1 dak 
Fastest Scam Ever 10 dak 
The Story of Bill Stephens | Fascinating Horror Shorts 1 dak 
The Cruise Industry's Arms Race 18 dak 
Name That Infrastructure (Ep. 5) #shorts 1 dak 
Practical Magic (1998) - Love Spell or True Romance? (9/11) | Movieclips 3 dak 
The Hidden Engineering Behind the Falkirk Wheel 16 dak 
Name That Infrastructure (Ep. 9) #shorts 1 dak 
The Weirdest Tool in Underwater Construction 18 dak 
How are holograms possible? 47 dak 
Name That Infrastructure (Ep. 12) #shorts 1 dak 
Why Are There No Short Arch Dams? 17 dak 
What is sumo, and why is it so popular? - Lee Thompson 6 dak 
How the Boeing 787 Works | Full Documentary 66 dak 
Plasma Speaker DIY (AM Demodulator) 11 dak 
We can't invent a robot better than these ferrets 22 dak 
The Bizarre Bases of Antenna Towers 19 dak 
Lindsey Clancy and the surprisingly simple solution to helping moms with postpartum depression 2 dak 
Should you rinse your dishes before placing them in the dishwasher? - Rachel Yang 6 dak 
Member Update "WHERES THE VIDEOS FURZE" 5 dak 
How The Space Shuttle Worked | Full Documentary 78 dak 
Why do gay people sound like that? 6 dak 
These Plants Have Digital Twins #science #scishow #plant #xray 1 dak 
Intro to Geology: Crash Course Geology #1 10 dak 
For all the hopeless romantics 2 dak 
Do cranberries help prevent UTIs? #shorts #science #stem #scishow 1 dak 
The Hidden Engineering of Runways 19 dak 
How Liquid Dampers in Skyscrapers Work 20 dak 
THE POWER SYSTEMS! | Real Life Power Armor (Part 2/6) 30 dak 
Tallest Sunflower EVER! 🌻 #PeppaPig #Shorts #Summer #LEGODUPLO 1 dak 
You Need to Get Your CCNA Right Now!! 10 dak 
Cruise Party | Safety Rules for Kids | Good Habits | Sheriff Labrador | Kids Cartoon | BabyBus 29 dak 
This Glove Punches For You! (COLD GAS ROCKET FIST) 20 dak 
The PROBLEM with Capitalism - Smarter Every Day 316 56 dak 
What’s really happening in the ocean’s “dark zones” 32 dak 
The Dark Truth of the Panama Canal 41 dak 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