MaskBit

Embedding-free Image Generation via Bit Tokens

1. We study the key ingredients of recent closed-source VQGAN tokenizers and develop a publicly available, reproducible, and high-performing VQGAN model, called VQGAN+, achieving a significant improvement of 6.28 rFID over… pic.twitter.com/NoTpp86WDZ

— AK (@_akhaliq) 25 septembre 2024

MaskBit Embedding-free Image Generation via Bit Tokens 1. We study the key ingredients of recent closed-source VQGAN tokenizers and develop a publicly available, reproducible, and high-performing VQGAN model, called VQGAN+, achieving a significant improvement of 6.28 rFID over

→ View original post on X — @_akhaliq