Back to News
Allen Institute for AI (Ai2)•December 10, 2025
Olmo-3.1 32B Instruct
Open Source
open-source
llm
reasoning
conversational-ai
### TL;DR
Olmo-3.1 32B Instruct is a powerful open-language model from the Olmo 3 family, specifically designed for conversational and reasoning tasks. It is pre-trained on the Dolma 3 dataset and post-trained on Dolci datasets, offering high-level performance across math, coding, and general reasoning benchmarks.
Key Insights & Metrics
Pricing
Free
Cost structure
Version
3.1
Current release version
Hardware
A high-end GPU with at least 24GB of VRAM for a compressed version, scaling up to an 80GB enterprise GPU (like an A100) for full precision, backed by 32GB+ of system RAM
Compute requirements
Category
Open Source
Licensing model
Region
United States
Primary region
Key Features
- 32-billion parameter autoregressive transformer architecture
- Enhanced reasoning capabilities for complex math and coding tasks
- Permissive Apache 2.0 licensing for research and commercial use
- Seamless integration with Transformers, vLLM, and SGLang
- Trained on high-quality Dolma 3 and Dolci-Instruct-RL datasets
Discussion
0
Upvotes
0
Downvotes
0 reviews
Sign in to leave a review
Reviews
No reviews yet. Be the first to review!