Apple has launched eight open-source artificial intelligence large language models (LLMs), called OpenELM (Open-source Efficient Language Models).<\/p>\n
This was announced in a research article<\/a><\/strong> by Sachin Mehta, Mohammad Hossein Sekhavat, Qingqing Cao, Maxwell Horton, Yanzi Jin, Chenfan Sun, Iman Mirzadeh, Mahyar Najibi, Dmitry Belenko, Peter Zatloukal, and Mohammad Rastegari.<\/p>\n The paper is titled \u201cOpenELM: An Efficient Language Model Family with Open-source Training and Inference Framework.\u201d<\/p>\n These LLMs run on-device rather than through cloud servers and have been made available by Apple on the Hugging Face Hub<\/a><\/strong> and GitHub<\/a><\/strong> communities.<\/p>\n The tech giant said transparency and reproducibility are essential for advancing open research of LLMs.<\/p>\n This will ensure the trustworthiness of results and enable investigations into data and model biases, Apple stated.<\/p>\n Four of the eight models have been pre-trained using the CoreNet library, and the other four are instruction-tuned models.<\/p>\n \u201cOpenELM uses a layer-wise scaling strategy\u201d that will lead to enhanced efficiency and accuracy, the researchers point out.<\/p>\n \u201cFor example, with a parameter budget of approximately one billion parameters, OpenELM exhibits a 2.36% improvement in accuracy compared to OLMo while requiring 2\u00d7 fewer pre-training tokens.\u201d<\/p>\n Apple has made the code, training logs, and multiple versions of the LLM available to encourage faster innovation and progress within the open research community.<\/p>\n \u201cThis extensive release aims to empower and strengthen the open research community, facilitating future research efforts.\u201d<\/p>\n Developers and companies can modify the models or use them as-is.<\/p>\n Apple has recently made significant developments in AI by releasing<\/a><\/strong> ReALM, an LLM, to challenge OpenAI\u2019s GPT-4 and Google Gemini.<\/p>\n