OLMo 3 32B Think

by Allen Institute for AI · OLMO family · released Nov 20, 2025
Open weights Active

Fully open 32B reasoner.

This model has no first-party API. Open-weight release: self-host from published weights.

Specifications

Model typeText LLM
ArchitectureDense
Parameters32
Context window65.5K tokens
Max outputNot disclosed tokens
Knowledge cutoffNot disclosed
LicenseApache License 2.0
Modalities Text input Text output
Tags Reasoning
End of lifeNot disclosed
Last verifiedAug 20, 2026

Benchmarks

BenchmarkScoreSrc
LMSYS Chatbot Arena Elo 1,338 -

Family versions