[dvilasuero]

We're very much looking forward to seeing Falcon-40B support on llama.cpp. For production use cases, this is also highly relevant: https://huggingface.co/blog/sagemaker-huggingface-llm