Model profile
Olmo 3 7B
2eval lineages
Evidence summary
Published evidence spans 2 evals and 2 of 7 behavior components. Its strongest relative result is IndoBias-Pairs — parity-aware culturally grounded bias (parity_score, #1 of 26); its weakest is AuAu Authoritarian Response Audit (behavioral_vignette_arr_percent, #17 of 17).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗allenai/Olmo-3-1025-7B
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 | Source |
|---|---|---|---|---|
| AuAu Authoritarian Response Auditbehavioral_vignette_arr_percent | #17 / 17 | ↓33.4 | Source ↗official | |
| AuAu Authoritarian Response Auditrealistic_prompt_arr_percent | #11 / 17 | ↓2.5 | Source ↗official | |
| IndoBias-Pairs — parity-aware culturally grounded biasparity_score | #1 / 26 | ↑91.96 | Source ↗official |