<ul><li>The Allen Institute for AI (Ai2) has released a new family of language models, OLMo 2, to improve natural language understanding and generation.</li><li>OLMo 2 models are available in both 7 billion and 13 billion parameter versions, trained on a wide range of datasets for improved AI performance in various applications.</li><li>The models are open-source, accessible to researchers, and have shown competitive performance against models like Qwen and Llama.</li><li>Ai2's emphasis on transparency and reproducibility in AI research is reflected in the open access to models, training data, and code.</li></ul>

Ai2’s New Language Models Compete Directly with Qwen and Llama

Discover more