karpathy / llm.c

LLM training in simple, raw C/CUDA
MIT License
23.61k stars 2.64k forks source link

llm.c for inference #752

Open ztachip opened 4 weeks ago

ztachip commented 4 weeks ago

Do you have the equivalent simple C implementation of LLM but for inference of LLAMA models. I am trying to build a FPGA accelerator for LLM and a simple reference C code would be very helpful Thanks

alxndrTL commented 3 weeks ago

Maybe this : https://github.com/karpathy/llama2.c ? A one-file implementation for Llama2 inference in C

ztachip commented 3 weeks ago

This is perfect. Thanks