Diffusion models are a class of generative models that create data by learning to reverse a gradual noising process, transforming random noise into structured outputs through iterative denoising. Operating on the principle of stochastic differential equations and score-based modeling, they have achieved state-of-the-art results in image, video, and audio generation. Unlike GANs which require adversarial training or VAEs which compress data into fixed latent spaces, diffusion models iteratively refine noise using learned score functions, enabling highly controllable generation with stable training dynamics. The key insight is that reversing a carefully designed forward diffusion process—which progressively adds noise—can be learned as a denoising task, allowing models to generate diverse, high-fidelity outputs by starting from pure noise and gradually removing it through multiple timesteps.