This only example from Unsloth.ai

This model based on Llama 3.2 3B is trained for understand the "Byte Latent Transformer: Patches Scale Better Than Tokens" [research paper]

(https://ai.meta.com/research/publications/## byte-latent-transformer-patches-scale-better-than-tokens/) that was ## published in December 2024.

This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.