preprint · ArXiv.org · 2025
Olmo 3
Team Olmo, :, Allyson Ettinger, Amanda Bertsch, Bailey Kuehl, David J. Graham, David Heineman, Dirk Groeneveld, Faeze Brahman, Finbarr Timbers, Hamish Ivison, Jacob Morrison, Jake Poznanski, Kyle Lo, Luca Soldaini, Matt Jordan, Mayee Chen, Michael Noukhovitch, Nathan Lambert, P N Walsh, Pradeep Dasigi, Robert Berry, Saumya Malik, Saurabh A. Shah, Scott Geng, Shane Arora, Shashank Gupta, Thomas Anderson, Tingting Xiao, Tyler Murray, Tyler Romero, Victoria Graf, Akari Asai, Akshita Bhagia, Alexander Wettig, Alisa Liu, Aman Rangapur, Chloe Anastasiades, Chia‐Hua Huang, Dustin Schwenk, Harsh Trivedi, Ian Magnusson, Jaron Lochner, Jia‐Cheng Liu, Lester James V. Miranda, Maarten Sap, Marsha Y. Morgan, Michael Schmitz, Michal Guerquin, Michael D. Wilson and 19 more
Cited by 0Open access
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.
Cite this paper
Olmo, T., :, Ettinger, A., Bertsch, A., Kuehl, B., Graham, D. J., Heineman, D., Groeneveld, D., Brahman, F., Timbers, F., Ivison, H., Morrison, J., Poznanski, J., Lo, K., Soldaini, L., Jordan, M., Chen, M., Noukhovitch, M., Lambert, N., . . . Wilson, M. D. (2025). Olmo 3. ArXiv.org. https://doi.org/10.48550/arxiv.2512.13961
Olmo, Team, et al. “Olmo 3.” ArXiv.org, 2025, https://doi.org/10.48550/arxiv.2512.13961.
Olmo, Team, :, Allyson Ettinger, Amanda Bertsch, Bailey Kuehl, David J. Graham, David Heineman, et al. 2025. “Olmo 3.” ArXiv.org. https://doi.org/10.48550/arxiv.2512.13961.
Olmo, T. et al. (2025) ‘Olmo 3’, ArXiv.org. Available at: https://doi.org/10.48550/arxiv.2512.13961.
@misc{olmo2025olmo,
title = {{Olmo 3}},
author = {Olmo, Team and : and Ettinger, Allyson and Bertsch, Amanda and Kuehl, Bailey and Graham, David J. and Heineman, David and Groeneveld, Dirk and Brahman, Faeze and Timbers, Finbarr and Ivison, Hamish and Morrison, Jacob and Poznanski, Jake and Lo, Kyle and Soldaini, Luca and Jordan, Matt and Chen, Mayee and Noukhovitch, Michael and Lambert, Nathan and Walsh, P N and Dasigi, Pradeep and Berry, Robert and Malik, Saumya and Shah, Saurabh A. and Geng, Scott and Arora, Shane and Gupta, Shashank and Anderson, Thomas and Xiao, Tingting and Murray, Tyler and Romero, Tyler and Graf, Victoria and Asai, Akari and Bhagia, Akshita and Wettig, Alexander and Liu, Alisa and Rangapur, Aman and Anastasiades, Chloe and Huang, Chia‐Hua and Schwenk, Dustin and Trivedi, Harsh and Magnusson, Ian and Lochner, Jaron and Liu, Jia‐Cheng and Miranda, Lester James V. and Sap, Maarten and Morgan, Marsha Y. and Schmitz, Michael and Guerquin, Michal and Wilson, Michael D. and others},
journal = {ArXiv.org},
year = {2025},
doi = {10.48550/arxiv.2512.13961}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.