Olmo 3.1 32B
Olmo 3.1 32B is an open-weight language model from Allen Institute for AI (Ai2), released on December 12, 2025.
- Developer
- Allen Institute for AI (Ai2)
- Released
- December 12, 2025
- Total parameters
- 32B
- Architecture
- Dense
- Context window
- 65K
- Input
- Text
- License
- Apache 2.0
- Commercial use
- Yes
- Official weights, GB
- 64.5 GB
- Hugging Face repository
- huggingface.co
- Checked on
Olmo 3.1 32B is a dense model with 32B parameters. Ai2 lists a context window of about 65K tokens. The model takes text input. The weights are released under Apache 2.0, which allows commercial use. The official weights on Hugging Face take about 64.5 GB in BF16. Running the model needs at least that much memory across GPUs and system RAM, plus room for the context cache; quantized versions need less.
- Total parameters, billions
- 32B
- License type
- Apache 2.0
- License conditions
- Apache 2.0. Ai2 also publishes Responsible Use Guidelines, which are guidance rather than license terms.
- Weights precision
- BF16
- License text
- apache.org
- Sources
Page URL Hugging Face model card https://huggingface.co/allenai/Olmo-3.1-32B-Instruct/raw/main/README.md Hugging Face model page https://huggingface.co/allenai/Olmo-3.1-32B-Instruct Hugging Face API https://huggingface.co/api/models/allenai/Olmo-3.1-32B-Instruct?expand=safetensors allenai.org announcement https://allenai.org/blog/olmo3
Questions
What license does Olmo 3.1 32B use?
How much memory does Olmo 3.1 32B need?
How many parameters does Olmo 3.1 32B have?
What context window does Olmo 3.1 32B have?
Sources
Every figure comes from the developer’s model card, license text and official announcements, and from the model’s Hugging Face repository metadata. Each model page lists its sources and the date they were checked; the register is rechecked every week.