Ai2: Olmo 3.1 32B Think
Model Type
Open Weight Model
32B parameters
Recommended Use Cases
Text Generation
The Think model from the OLMo 3 model family — a fully open 32B parameter reasoning model from Ai2 (Allen Institute for AI). Unlike other open LLMs, OLMo provides complete access to training data, code, checkpoints, and logs.
Per its creators, Ai2:
OLMo 3.1 32B Think is capable of reasoning through complex problems step by step. A strong platform for RL research and other advanced experiments that need serious horsepower.
Long chain-of-thought thinking improves reasoning tasks like math and coding.
Key Capabilities
- Math: 96.2% on MATH, 80.6% on AIME 2024, 78.1% on AIME 2025
- Coding: 91.5% on HumanEvalPlus, 83.3% on LiveCodeBench v3
- Reasoning: 88.6% on BigBenchHard, 80.1% on ZebraLogic