Dockerfile

# File: Dockerfile
FROM ghcr.io/ggerganov/llama.cpp:server

COPY app/Llama-3-8B-Instruct-v0.1.Q4_K_M.gguf /models/

CMD ["/app/llama-server", "--model", "/models/Llama-3-8B-Instruct-v0.1.Q4_K_M.gguf", "--alias", "llama3", "--ctx-size", "2048"]