Diffusion Models

• They are Text To Image models
• Trains using a forward diffusion process
• Adds noise to the image
• Generates image from noise
• Reverse the diffusion process

Providers

• Model: Stable Diffusion, Provider/Company: Stability AI
• Model: Midjourney, Provider/Company: Midjourney https://www.midjourney.com/ (got famous on discord)
• Model: Dall-E, Provider/Company: OpenAI