metadata
license: llama3
This is converted Llama 3 8B model to Distributed Llama format. The model is quantized to Q40.
π How to Run?
- β¬ Download this repository.
- β¬ Download Distributed Llama repository.
- π¨ Build Distributed Llama:
make dllama
- π Run Distributed Llama:
./dllama inference --prompt "Hello world" --model dllama_model_meta-llama-3-8b_q40.m --tokenizer dllama_tokenizer_llama3.t --buffer-float-type q80 --nthreads 4