European Open Source AI Index

Database News Guides About Contribute

OLMo

by Ai2

About the model:

Open LLM trained from scratch by Allen AI.

Model type:

Text

Model performance class:

Full

Link to the model:

https://huggingface.co/allenai/Olmo-3.1-32B-Think

Base models:

Olmo-3-1125-32B

End model:

Olmo-3.1-32B-Think

End model license:

Apache-2.0

About the organisation:

Allen Institute for AI (non-profit research institute)

Link to the organisation:

https://allenai.org

Model release date:

December 2025

Availability

Base Model Data

Training data for base model released and use documented.

https://huggingface.co/datasets/allenai/dolma3_mix-5.5T-1125 https://huggingface.co/datasets/allenai/dolma3_longmino_mix-100B-1125 https://huggingface.co/datasets/allenai/dolma3_dolmino_mix-100B-1125

End User Model Data

Data for fine-tuning published in a well-organized manner.

https://huggingface.co/datasets/allenai/Dolci-Think-RL-32B

Base Model Weights

Model weights made available on HuggingFace.

https://huggingface.co/allenai/Olmo-3-1125-32B

End User Model Weights

Model weights made available on HuggingFace.

https://huggingface.co/allenai/Olmo-3.1-32B-Think

Training Code

Multiple repos with training, architecture and fine-tuning code available.

https://github.com/allenai/open-instruct https://github.com/allenai/OLMo-core

Documentation

Code Documentation

Repositories and code well-described, commented and documented.

https://github.com/allenai/open-instruct https://github.com/allenai/OLMo-core

Hardware Architecture

Architecture documented in requisite detail.

https://aclanthology.org/2024.acl-long.841/

Preprint

Pre-print goes into impressive detail about the data, training process, architecture, and evaluation.

https://huggingface.co/papers/2512.13961

Paper

Conference paper published in COLM 2025. More under review?

https://openreview.net/forum?id=2ezugTT9kU#discussion

Modelcard

Model card provides broad overview and links to full details.

https://huggingface.co/allenai/Olmo-3.1-32B-Think

Datasheet

Data sheets are well-documented and provide requisite info.

https://huggingface.co/datasets/allenai/dolma3_mix-5.5T-1125 https://huggingface.co/datasets/allenai/dolma3_longmino_mix-100B-1125 https://huggingface.co/datasets/allenai/dolma3_dolmino_mix-100B-1125 https://huggingface.co/datasets/allenai/Dolci-Think-RL-32B https://huggingface.co/datasets/allenai/Dolci-Instruct-RL-7B https://huggingface.co/datasets/allenai/Dolci-Think-DPO-32B https://huggingface.co/datasets/allenai/Dolci-Instruct-DPO-7B https://huggingface.co/datasets/allenai/Dolci-Think-SFT-32B https://huggingface.co/datasets/allenai/Dolci-Instruct-SFT-7B

Access

Licenses

Apache 2.0, an OSI-approved license.

https://huggingface.co/allenai/Olmo-3.1-32B-Think#model-description

Is this information not up to date?

Contribute here ->

Supported by the Centre for Language Studies and the Dutch Research Council. Website design & development © 2024 by BSTN. This version of the index generated 09 April 2026, website content last updated 11 March 2026.