LLM Reference

OLMo 3 32B Think

Released
2026-03-01
Last refreshed
2026-04-27
Status
Researched 45d ago
Open SourceCommercial use allowedCodingClassification

OLMo 3 32B Think has model metadata, but missing tracked provider pricing keeps it from being a default production pick.

Use it for

  • Teams evaluating coding and classification
  • Workloads that can use a 64k context window

Do not use it for

  • Cost-sensitive launches that need sourced token pricing
  • Vision or document-understanding workloads
  • Strict JSON or tool-calling flows
Specifications
Family
OLMo
Released
2026-03-01
Context
64k
Parameters
32B
Architecture
Decoder Only
Knowledge cutoff
2024-12
Specialization
general
Openness
Open source
License
Apache 2.0(OSI)Commercial use allowed
Training
finetuned
Created by
Allen Institute for Artificial Intelligence (AI2)Allen Institute for Artificial Intelligence (AI2)

Advocating for open science and source

Seattle, Washington, United States
Founded 2014
Website
Pricing

No tracked provider token pricing is available yet.

Links
Website HuggingFace

About

OLMo 3 32B Think is Allen Institute for AI's reasoning-focused model with extended thinking chains for complex logic problems and multi-step reasoning.

OLMo 3 32B Think is an open-source model in the OLMo family. The structured metadata tracks a 64k-token context window and reasoning. Headline tracked benchmarks include AIME 2024 76.8, AIME 2025 72.5, and Google-Proof Q&A 58.1.

Top use-case fit: coding, agents, and build tasks

Coding

2 relevant benchmarks in the decision map.

Classification

1 relevant benchmark in the decision map.

Provider price ladder

No tracked provider token pricing is available for this model yet.

Capabilities

Reasoning

Benchmark peer barsfor Coding

HumanEvalRank 16 of 96
Claude Sonnet 4.6
98.0
o3
96.7
Claude Opus 4.6
95.0
Grok-3
94.5
OLMo 3 32B Thinkcurrent
91.4
LiveCodeBenchRank 18 of 52
DeepSeek V4 Pro
93.5
Gemini 3.1 Pro Preview
91.7
DeepSeek V4 Flash
91.6
Qwen3.7-Max
91.6
OLMo 3 32B Thinkcurrent
83.5

Benchmark scores(7)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionSource
AIME 202476.8AIME 2024 (accuracy)https://huggingface.co/allenai/Olmo-3-32B-Think
AIME 202572.5AIME 2025 (accuracy)https://huggingface.co/allenai/Olmo-3-32B-Think
Google-Proof Q&A58.1GPQA (accuracy)https://huggingface.co/allenai/Olmo-3-32B-Think
HumanEval91.4HumanEval (pass@1)https://huggingface.co/allenai/Olmo-3-32B-Think
LiveCodeBench83.5LiveCodeBench v3 (accuracy)https://huggingface.co/allenai/Olmo-3-32B-Think
MATH-50096.1MATH benchmark (accuracy)https://huggingface.co/allenai/Olmo-3-32B-Think
Massive Multitask Language Understanding85.4From official HuggingFace model card (accuracy)https://huggingface.co/allenai/Olmo-3-32B-Think

Migration checks

No linked migration route is available for this model yet.