AllenAI: Olmo 3 32B Think

AllenAI: Olmo 3 32B Think

allenai · Released Nov 21, 2025
35
Our Score

Olmo 3 32B Think is a large-scale, 32-billion-parameter model purpose-built for deep reasoning, complex logic chains and advanced instruction-following scenarios. Its capacity enables strong performance on demanding evaluation tasks and highly nuanced conversational reasoning. Developed by Ai2 under the Apache 2.0 license, Olmo 3 32B Think embodies the Olmo initiative’s commitment to openness, offering full transparency across weights, code and training methodology.

$0.15 / 1M Input Price
$0.50 / 1M Output Price
65,536 tokens Context Window
65,536 tokens Max Output
32B Parameters

Architecture

ModalityText → Text
TokenizerOther
Parameters32B

Performance Indices

Source: Artificial Analysis

12.1 Intelligence Index
10.5 Coding Index
1.5 Agentic Index
73.7 Math Index

Benchmark Scores

Evaluations

GPQA Diamond 61%
Graduate-level scientific reasoning
HLE 5.9%
Humanity's Last Exam
MMLU Pro 75.9%
Multi-task language understanding
LiveCodeBench 67.2%
Live coding evaluation
SciCode 28.6%
Scientific computing
AIME 2025 73.7%
Competition mathematics (2025)
IFBench 49.1%
Instruction following
TerminalBench Hard 1.5%
Agentic terminal tasks

Benchmark data from Artificial Analysis and Hugging Face

Model Information

OpenRouter ID allenai/olmo-3-32b-think
Providerallenai
Release Date November 21, 2025
Context Length65,536 tokens
Max Completion65,536 tokens
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.15 $0.000150
Output $0.50 $0.000500

Live Performance

Live endpoint metrics — refreshed every 30 minutes.

387ms
Best Latency (TTFT)
96.5 tok/s
Best Throughput
0/1
Active Endpoints
Available via: Parasail