Susunod
But how do AI images and videos actually work? | Guest video by Welch Labs 38 min 
How Disney Built A Realistic Lightsaber 17 min 
I Thought this would be Safe! 17 min 
Can Tech Transform my Body in 3 Months? 13 min 
Exploring the Paranoid Country with 374,142 Bunkers to Hide Everyone (how is this possible?) 35 min 
Bretagne 3 min 
The Polar Cruisers: Winter Climb to the Hanging Cities | AI 4K — Retrofutura 8 min 
I put high Speed Prop Blades on my Kayak Paddle!! 15 min 
Army of Fear 3 min 
100 Hours in Europe's Most Corrupt Country 34 min 
Blinded By A Rollercoaster: The Problem With Dragon Challenge | Fascinating Horror 10 min 
I'll Always Remember You 13 min 
Terrible Hikers 12 min 
The Terrible Design of Newark Airport 17 min 
This 3 day solo winter hike may well have changed my life... 70 min 
The Failed Logistics of Hurricane Katrina 21 min 
Dating Girls In Front Of Their Dads 14 min 
It’s not your typical jigsaw puzzle — it’s your next great adventure 1 min 
GUESS THE ACTOR FT IDRIS ELBA 45 min 
The Hidden Engineering of Runways 19 min 
The Hidden Engineering of Niagara Falls 16 min 
Woman Rescues Flooded Car And Restores It Back To New | By @Flame-Frame-VN 
The Bizarre Bases of Antenna Towers 19 min 
Name That Infrastructure (Ep. 3) #shorts 1 min 
Dad Builds 2 DIY Dream MOTORBIKES For His Daughters | Start To Finish 
Name That Infrastructure (Ep. 11) #shorts 1 min 
Name That Infrastructure (Ep. 1) #shorts 1 min 
Century III Mall 2025 Update 13 min 
What The Heck Is Tuberculosis? #science #scishow #medicine #johngreen 1 min 
Will gas prices decide the midterms? | America, Actually 28 min 
Lorn - Acid Rain (Official Music Video) 3 min 
Ever walk into a room and forget what you were doing? - Gabriel Radvansky 6 min 
The V-22’s prototype was even more dangerous 3 min 
How Long Does Sperm REALLY Live? #science #scishow #STEM #hankgreen 1 min 
The Shoe That Broke Running 16 min 
Behind-the-scenes look at illustrator Senne Trip creating the art of Pandora’s Legacy 1 min 
Crash Course Astronomy Marathon • Full Series 502 min 
What your nipple shape says about you 5 min 
Amtrak's Subtle, Surprising Success 20 min 
What Dinosaurs, Pyramids, and the Atomic Bomb Have in Common 2 min 
Crash Course Hosts Read Comments 2 min 
Name That Infrastructure (Ep. 8) #shorts 1 min 
The NFL's Concussion Crisis 22 min 
SciShow is heading into the field TOMORROW with the all new Field Trips mini series. 1 min 
Peppa Pig Paints The BIGGEST Picture in the World! 🎨 👩🎨 Peppa Pig Full Episodes | 1 Hour 63 min 
Why do some people have a better sense of direction than others? - Oliver Baumann 6 min 
1955 vs 2025, who actually had it better? 25 min 
How billionaires buy American elections 45 min 
Red gold: The world's most expensive spice - Carolyn Beans 7 min 
AI News: An INSANE Week… Here’s What Matters 31 min 
Which of these is the best idea? 🤷♂️ #batmobile #batman #engineering 1 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