[Refresh for a random deep learning StyleGAN-generated anime face & GPT-2-small-generated anime plot; reloads every 15s. For many waifus simultaneously in a randomized grid, see "These Waifus Do Not Exist". A partial image archive is available for download.
Technical note: these faces were generated with Ψ=0.7 for images #0–60,000 (high quality but low diversity), Ψ=1.0 for #60,001–70,000 (low quality, high diversity), and Ψ=0.6 for #70,001–100,000 for a mix of good & interesting faces; for details about the Ψ hyperparameter, see the StyleGAN paper. The "plot summaries" were generated by the public GPT-2-small. For more details, see the TWDNE site implementation writeup.]
