With Speculative Sampling (SSp), a large language model can generate tokens quite faster using a smaller model as help. This repo shows how it's done and measures timing improvements using Llama ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results