Ai2 - Dense - verified
Run Ai2 Olmo 3.1 32B Think locally
Ai2 Olmo 3.1 32B Think is a 32.2B dense model from Ai2 with a 66K-token context window, released under Apache 2.0 (commercial use is allowed); the smallest listed download is 18.1 GB, the 4bit MLX build from mlx-community/Olmo-3.1-32B-Think-4bit.
Official weights: allenai/Olmo-3.1-32B-Think
- Total params
- 32.2B
- Architecture
- Dense
- Context
- 66K
- Smallest download
- 18.1 GB
Fits in
Smallest download at most 85% of memoryDownloads
The official weights and every quantization listed for this model, with file sizes from Hugging Face.
| Download | Format | Size | Repo |
|---|---|---|---|
Official weights | BF16 | 64.5 GB | allenai/Olmo-3.1-32B-Think |
Q4_K_M | GGUF | 19.5 GB | bartowski/allenai_Olmo-3.1-32B-Think-GGUF |
Q8_0 | GGUF | 34.3 GB | bartowski/allenai_Olmo-3.1-32B-Think-GGUF |
4bitSmallest listed | MLX | 18.1 GB | mlx-community/Olmo-3.1-32B-Think-4bit |
GB is decimal: 1 GB = 10^9 bytes. Add memory for the context and the operating system on top of the file size.
Run it
Copied verbatim from the page linked under each command. Server commands from model cards often assume a multi-GPU node; check the flags before you run them.
Not sure which runtime to use? Compare local LLM runtimes by platform, API and features.
License
- License
- Apache 2.0
apache-2.0 - Commercial use
- Commercial use: YesCommercial use is allowed.
- Gated download
- No. The files download directly.
Worth knowing
Requires transformers 4.57.0 or newer; the card gives no context length, so the figure is config.json max_position_embeddings.
Released (Hugging Face repo creation date).
FAQ
How big is the Ai2 Olmo 3.1 32B Think download?
The official BF16 weights in allenai/Olmo-3.1-32B-Think are 64.5 GB. The smallest listed download is 18.1 GB, the 4bit MLX build from mlx-community/Olmo-3.1-32B-Think-4bit. Plan for more memory than the file size: the context (KV cache) and the operating system need room too.
Can I use Ai2 Olmo 3.1 32B Think commercially?
Ai2 Olmo 3.1 32B Think is released under Apache 2.0, and commercial use is allowed. The license text is at https://www.apache.org/licenses/LICENSE-2.0.
How much memory do I need for Ai2 Olmo 3.1 32B Think?
By this directory's rule of thumb it fits in 24 GB and up: the smallest listed download, 18.1 GB, is at most 85% of 24 GB, which leaves room for the context and the operating system.
How do I run Ai2 Olmo 3.1 32B Think with Ollama?
Run "ollama run olmo-3.1:32b-think", as shown at https://ollama.com/library/olmo-3.1:32b-think. The Ollama library lists the olmo-3.1:32b-think tag at 19GB.
What is the context length of Ai2 Olmo 3.1 32B Think?
65,536 tokens. A longer context needs more memory for the KV cache, so you can run with a shorter one when memory is tight.
Sources
Last verified from the pages below.
- HF API (params, license, createdAt) - huggingface.co
- HF file tree - huggingface.co
- Model card - huggingface.co
- config.json - huggingface.co
- Ollama tags - ollama.com
- GGUF tree (bartowski) - huggingface.co
- MLX tree - huggingface.co
Newsletter
New open weights, every week
One email per week. New model releases, open-weight drops and what they mean for what you build.
Keep exploring
Best local AI models
Our curated picks by hardware tier, with benchmarks and the reasoning.
Model API pricing and specs
First-party prices per 1M tokens when you would rather call an API.
AI benchmark leaderboards
Sourced scores, for when you need to compare quality.
Setup guides
Point your coding tools at the model you picked.