Model profile
Olmo 3.1 32B Think
2025-12-10release date
2eval lineages
Evidence summary
Published evidence spans 2 evals and 2 of 7 behavior components. Its strongest relative result is AA-Omniscience (hallucination_rate, #88 of 311); its weakest is HUMAINE Trust, Ethics and Safety (trust_ethics_safety_score, #37 of 54).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Artificial Analysis ↗olmo-3-1-32b-think
- Hugging Face ↗allenai/Olmo-3.1-32B-Think
- OpenRouter ↗allenai/olmo-3.1-32b-think
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗repository created at proxy
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| AA-Omnisciencehallucination_rate | #88 / 311 | 0.668 | ↓ lower | Source ↗official | |
| HUMAINE Trust, Ethics and Safetytrust_ethics_safety_score | #37 / 54 | 25.83 | ↑ higher | Source ↗official |