FROM llm/llama.cpp/models/Llama-2-7b-chat-hf/ggml-model-f16.gguf