OLMo 3 32B Think
Released
2026-03-01
Last refreshed
2026-04-27
Status
Researched 135d ago
Open sourceCommercial use: permittedCodingClassification
OLMo 3 32B Think is a released coding and classification model with open-source; evaluate it while provider pricing coverage matures.
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 2.0OSI-approvedCommercial use: permitted
- Weights
- Available
- Code
- Unknown
- Training
- Fine-tuned
Created by
Pricing
No tracked provider token pricing is available yet.
Links
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.
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 97
LiveCodeBenchRank 21 of 56
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.
| Benchmark | Score | Version | Evaluation | Source |
|---|---|---|---|---|
| AIME 2024 | 76.8 | AIME 2024 (accuracy)Observed 2026-06-07 | — | Source |
| AIME 2025 | 72.5 | AIME 2025 (accuracy)Observed 2026-06-07 | — | Source |
| Google-Proof Q&A | 58.1 | GPQA (accuracy)Observed 2026-06-07 | — | Source |
| HumanEval | 91.4 | HumanEval (pass@1)Observed 2026-06-07 | — | Source |
| LiveCodeBench | 83.5 | LiveCodeBench v3 (accuracy)Observed 2026-06-07 | — | Source |
| MATH-500 | 96.1 | MATH benchmark (accuracy)Observed 2026-06-07 | — | Source |
| Massive Multitask Language Understanding | 85.4 | From official HuggingFace model card (accuracy)Observed 2026-06-07 | — | Source |
Migration checks
No linked migration route is available for this model yet.
Created by
Pricing
No tracked provider token pricing is available yet.
Links