Year: 2025
Type: preprint
Abstract: We introduce Olmo 3, a family of state-of-the-art, fully-open language models at the 7B and 32B parameter scales. Olmo 3 model construction targets long-context reasoning, function calling, coding, instruction following, general chat, and knowledge recall. This release includes the entire model flow, i.e., the full lifecycle of the family of models, including every stage, checkpoint, data point, and dependency used to build it. Our flagship model, Olmo 3 Think 32B, is the strongest fully-open thinking model released to-date.
Source: arXiv (Cornell University)
Authors: Olmo, Team, :, Ettinger, Allyson, Bertsch, Amanda, Kuehl, Bailey, Graham, David, Heineman, David, Groeneveld, Dirk, Brahman, Faeze, Timbers, Finbarr, Ivison, Hamish, Morrison, Jacob, Poznanski, Jake, Lo, Kyle, Soldaini, Luca, Jordan, Matt, Chen, Mayee, Noukhovitch, Michael, Lambert, Nathan, Walsh, Pete, Dasigi, Pradeep, Berry, Robert, Malik, Saumya, Shah, Saurabh, Geng, Scott, Arora, Shane, Gupta, Shashank, Anderson, Taira, Xiao, Teng, Murray, Tyler, Romero, Tyler, Graf, Victoria, Asai, Akari, Bhagia, Akshita, Wettig, Alexander, Liu, Alisa, Rangapur, Aman, Anastasiades, Chloe, Huang, Costa, Schwenk, Dustin, Trivedi, Harsh, Magnusson, Ian, Lochner, Jaron, Liu, Jiacheng, Miranda, Lester James V., Sap, Maarten, Morgan, Malia, Schmitz, Michael, Guerquin, Michal, Wilson, Michael, Huff, Regan, Bras, Ronan Le, Xin, Rui, Shao, Rulin, Skjonsberg, Sam, Shen, Shannon Zejiang, Li, Shuyue Stella, Wilde, Tucker, Pyatkin, Valentina, Merrill, Will, Chang, Yapei, Gu, Yuling, Zeng, Zhiyuan, Sabharwal, Ashish, Zettlemoyer, Luke, Koh, Pang Wei, Farhadi, Ali, Smith, Noah A., Hajishirzi, Hannaneh
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4417465430