TLDR · QUICK LINKS
SAYS STUDY
Summary from the TLDR newsletter
Chinchilla 70B, a large language model by DeepMind, can perform lossless compression on images to 43.4% of their original size and audio data to just 16.4% of their raw size.
TLDR · QUICK LINKS
Chinchilla 70B, a large language model by DeepMind, can perform lossless compression on images to 43.4% of their original size and audio data to just 16.4% of their raw size.