cgbur/llama2.zig

This is stars and forks stats for /cgbur/llama2.zig repository. As of 02 May, 2024 this repository has 68 stars and 3 forks.

llama2.zig This is the Zig version of llama2.c by Andrej Karpathy. It runs inference for the llama2 model architecture recently published by Meta. It currently supports: Inference of llama2 model checkpoints Temperature control Top-p (nucleus) sampling Prompt handling (bpe tokenization) Sequence length control Custom tokenizers Multiquery support Running really fast The ultimate goal is to create a fast, portable, and user-friendly implementation of the llama2 model architecture. The code prioritizes...