[{"data":1,"prerenderedAt":79},["ShallowReactive",2],{"term-d\u002Fdiffusion-model":3,"related-d\u002Fdiffusion-model":60},{"id":4,"title":5,"acronym":6,"body":7,"category":40,"description":41,"difficulty":42,"extension":43,"letter":44,"meta":45,"navigation":46,"path":47,"related":48,"seo":54,"sitemap":55,"stem":58,"subcategory":6,"__hash__":59},"terms\u002Fterms\u002Fd\u002Fdiffusion-model.md","Diffusion Model",null,{"type":8,"value":9,"toc":33},"minimark",[10,15,19,23,26,30],[11,12,14],"h2",{"id":13},"eli5-the-vibe-check","ELI5 — The Vibe Check",[16,17,18],"p",{},"Diffusion models generate images by learning to reverse noise. In training, you take an image and slowly add random noise until it's pure static. The model learns to reverse this — given static, reconstruct the original. At generation time, start with pure noise, and the model gradually removes noise until a coherent image appears. That's how Stable Diffusion and DALL-E work.",[11,20,22],{"id":21},"real-talk","Real Talk",[16,24,25],{},"Diffusion models are generative models that learn to reverse a noise diffusion process. During training, Gaussian noise is progressively added to data (forward process). A neural network (U-Net or transformer) learns to predict and remove the noise (reverse process). At inference time, samples are generated by iteratively denoising from pure Gaussian noise. Stable Diffusion, DALL-E 3, and Sora are based on this approach.",[11,27,29],{"id":28},"when-youll-hear-this","When You'll Hear This",[16,31,32],{},"\"Stable Diffusion is a diffusion model for image generation.\" \u002F \"Diffusion models outperformed GANs on image quality benchmarks.\"",{"title":34,"searchDepth":35,"depth":35,"links":36},"",2,[37,38,39],{"id":13,"depth":35,"text":14},{"id":21,"depth":35,"text":22},{"id":28,"depth":35,"text":29},"ai","Diffusion models generate images by learning to reverse noise. In training, you take an image and slowly add random noise until it's pure static.","advanced","md","d",{},true,"\u002Fterms\u002Fd\u002Fdiffusion-model",[49,50,51,52,53],"Generative AI","GAN","Computer Vision","Training","Inference",{"title":5,"description":41},{"changefreq":56,"priority":57},"weekly",0.7,"terms\u002Fd\u002Fdiffusion-model","PtB40d_IXRrDV_V8-r7pordlfwJrAGsUCHuG7SZfIOw",[61,65,69,72,76],{"title":51,"path":62,"acronym":6,"category":40,"difficulty":63,"description":64},"\u002Fterms\u002Fc\u002Fcomputer-vision","beginner","Computer Vision is teaching AI to understand images and video. How does your phone unlock with your face? Computer Vision.",{"title":50,"path":66,"acronym":67,"category":40,"difficulty":42,"description":68},"\u002Fterms\u002Fg\u002Fgan","Generative Adversarial Network","A GAN is two neural networks fighting each other. One (the Generator) tries to create fake images that look real.",{"title":49,"path":70,"acronym":6,"category":40,"difficulty":63,"description":71},"\u002Fterms\u002Fg\u002Fgenerative-ai","Generative AI is AI that creates new stuff — text, images, code, music, video — rather than just classifying or predicting. ChatGPT writes essays.",{"title":53,"path":73,"acronym":6,"category":40,"difficulty":74,"description":75},"\u002Fterms\u002Fi\u002Finference","intermediate","Inference is when the AI actually runs and generates output — as opposed to training, which is when it's learning.",{"title":52,"path":77,"acronym":6,"category":40,"difficulty":74,"description":78},"\u002Fterms\u002Ft\u002Ftraining","Training is the long, expensive process where an AI learns from data.",1776518274517]