Finetune Llama 3, Mistral & Gemma LLMs 2-5x faster with 80% less memory
Primary LanguagePythonApache License 2.0Apache-2.0