GitHub - facebookresearch/blt: Code for BLT research paper

We introduce the Byte Latent Transformer architecture (BLTs), a new byte-level LLM architecture that for the first time, matches tokenization-based LLM performance at scale, with significant improvements in inference efficiency and robustness. BLT encodes bytes into dynamically sized patches, which serve as the primary units of computation. Patches are segmented dynamically based on the entropy of