<ul><li>The Allen Institute for AI (AI2) has introduced OLMo 2 32B, a fully open model surpassing GPT-3.5 Turbo and GPT-4o mini on multi-skill benchmarks.</li><li>OLMo 2 32B has 32 billion parameters and achieved performance levels comparable to leading open-weight models with less computational resources.</li><li>It demonstrated impressive results on tasks like MMLU, MATH, and IFEval, showcasing its versatility in linguistic challenges.</li><li>By promoting openness and collaboration, AI2 enables researchers worldwide to build upon this work and contributes to the evolving landscape of AI.</li></ul>

Allen Institute for AI (AI2) Releases OLMo 32B: A Fully Open Model to Beat GPT 3.5 and GPT-4o mini on a Suite of Multi-Skill Benchmarks

Discover more