[2512.13961] Olmo 3

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.

Olmo 3 Olmo Team Allyson Ettinger 1 Amanda Bertsch 1, 3 Bailey Kuehl 1 David Graham 1 David Heineman 1 Dirk Groeneveld 1 Faeze Brahman 1 Finbarr Timbers 1 Hamish Ivison 1, 2 Jacob Morrison 1, 2 Jake Poznanski 1 Kyle Lo 1, 2 Luca Soldaini 1 Matt Jordan 1 Mayee Chen 1, 4 Michael Noukhovitch 1, 5, 6 Nathan Lambert 1 Pete Walsh 1 Pradeep Dasigi 1 Robert…

related reading

• Olmo 3: Charting a path through the model flow to lead open-source AI | Ai2allenai.org
• Olmo from Ai2allenai.org
• frontier model training methodologies | Alex Wa's Blogdjdumpling.github.io
• Qwen3: Think Deeper, Act Faster | Qwenqwenlm.github.io
• The Big LLM Architecture Comparisonmagazine.sebastianraschka.com
• [2005.14165] Language Models are Few-Shot Learnersarxiv.org
• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• 2403.09611.pdfarxiv.org
• DeepSeek-V3.2: Pushing the Frontier of Open Large Language Modelsarxiv.org
• Kimi K3 Tech Blog: Open Frontier Intelligencekimi.com
• 2310.10631arxiv.org
• 2305.14314arxiv.org