Stable diffusion dreams of steampunk brains

Andrej Karpathy · Intermediate ·📐 ML Fundamentals ·3y ago

Key Takeaways

The video demonstrates the use of Stable Diffusion to generate an ultrarealistic steam punk neural network machine in the shape of a brain, with dramatic lighting, using a noise vector as input and spherical interpolation to create a smooth animation. The video showcases the capabilities of Stable Diffusion in generating high-quality images from text prompts.

Full Transcript

[Music] [Music] do so [Music] [Applause] [Music] [Music] so [Music] so [Music] so [Music] [Applause] [Music] [Music] [Applause] [Music] [Music] so [Music] [Music] so [Music] [Music] so [Music] [Applause] [Music] [Music] [Music] so [Music] so [Music] [Music] [Applause] [Music] [Music] [Applause] [Music] [Music] so [Music] great [Music] so [Music] [Music] [Applause] [Music] [Music] [Music] so [Applause] [Music] [Music] [Applause] [Music] [Music] [Applause] [Music] [Music] [Music] [Music] [Music] so [Music] [Music] [Applause] [Music] so [Music] so [Music] so [Music] so [Applause] so [Music] [Music] [Applause] [Music] so [Music] [Applause] [Music] [Music] [Music] so [Music] [Music] so [Music] so [Music] [Applause] [Music] so [Music] [Music] [Music] so [Music] [Music] [Applause] so [Music] [Applause] [Music] [Music] [Music] so [Music] [Music] so [Music] [Music] [Applause] [Music] so [Music] [Music] you [Music] so [Music] so [Music] [Applause] [Music] [Music] [Applause] [Music] [Music] [Music] [Music] so [Music] [Music] [Applause] [Music] [Music] foreign [Music]

Original Description

Prompt: "ultrarealistic steam punk neural network machine in the shape of a brain, placed on a pedestal, covered with neurons made of gears. dramatic lighting. #unrealengine" Stable diffusion takes a noise vector as input and samples an image. To create this video I smoothly (spherically) interpolate between randomly chosen noise vectors and render frames along the way. This video was produced by one A100 GPU dreaming about the prompt overnight (~8 hours). While I slept and dreamt about other things. This is version 2 video of this prompt, with (I think?) a bit higher quality and trippy AGI music. Music: Wonders by JVNA Links: - Stable diffusion: https://stability.ai/blog - Code used to make this video: https://gist.github.com/karpathy/00103b0037c5aaea32fe1da1af553355 - My twitter: https://twitter.com/karpathy
Watch on YouTube ↗ (saves to browser)
Sign in to unlock AI tutor explanation · ⚡30

The video teaches how to use Stable Diffusion to generate high-quality images from text prompts, and how to create smooth animations using spherical interpolation. The key takeaway is that Stable Diffusion can be used to generate ultrarealistic images with dramatic lighting.

Key Takeaways
  1. Define a text prompt for the image to be generated
  2. Use Stable Diffusion to generate an image from the text prompt
  3. Apply spherical interpolation to create a smooth animation
  4. Refine the image using dramatic lighting
💡 Stable Diffusion can be used to generate high-quality images from text prompts, and spherical interpolation can be used to create smooth animations.

Related Reads

📰
Same Time Last Year, development was so different. This is what 10 years of programming taught me.
A developer reflects on 10 years of programming experience and shares key takeaways on how the field has evolved
Dev.to · SSK
📰
The best config in your bake-off didn't win. Selection did.
Learn how selection bias affects model evaluation and why the best config in your bake-off didn't win
Dev.to · Alexey Spinov
📰
Your AI query did not change. The Postgres plan did.
Learn how Postgres plan changes can impact AI query performance without altering the SQL query itself
Dev.to · Mads Hansen
📰
Python's Object Model in Depth: Why Two Lines That Look the Same Behave Differently
Understand Python's object model to explain differing behavior of two similar lines of code
Dev.to · Ameer Abdullah
Up next
The Adam Optimizer is Just Momentum + RMSProp
DataMListic
Watch →